MASTER NEGATIVE 
 
 NO. 93-81335- 
 
MICROFILMED 1993 
 COLUMBIA UNIVERSITY LIBRARIES/NEW YORK 
 
 as part of the 
 "Foundations of Western Civilization Preservation Project" 
 
 Funded by the 
 NATIONAL ENDOWMENT FOR THE HUMANITIES 
 
 Reproductions may not be made without permission from 
 
 Columbia University Library 
 
COPYRIGHT STATEMENT 
 
 The copyright law of the United States - Title 17, United 
 States Code - concerns the making of photocopies or 
 other reproductions of copyrighted material. 
 
 Under certain conditions specified in the law, libraries and 
 archives are authorized to furnish a photocopy or other 
 reproduction. One of these specified conditions is that the 
 photocopy or other reproduction is not to be "used for any 
 purpose other than private study, scholarship, or 
 research.** If a user makes a request for, or later uses, a 
 photocopy or reproduction for purposes in excess of **fair 
 use,** that user may be liable for copyright infringement. 
 
 This institution reserves the right to refuse to accept a 
 copy order if, in its judgement, fulfillment of the order 
 would involve violation of the copyright law. 
 
A UTHOR: 
 
 ELLIS, ROBERT 
 
 TITLE: 
 
 A TREATISE ON 
 HANNIBAL'S PASSAGE 
 
 PLACE: 
 
 CAMBRIDGE [ENG.] 
 
 DA TE: 
 
 1853 
 
COLUMBIA UNIVm^lTY LIBRARIES 
 PRESERVATION DEPARTMENT 
 
 DIBLIOCRAPHICMirROrORMTAKCFT 
 
 Master Negative // 
 
 Original Material as Filmed - Existing Bibliographic Record 
 
 874,04 
 ELS 
 
 ■ — * — ■■ . p j p» 
 
 -«i^r«« 
 
 "W ' tij— >jy^y 
 
 Ellis, Robert, 1820-1885, ;. 
 
 A treatise on nannibal's passage of thp Alna in t^].:^u 
 h;s ^ i« traced over th|Lit?)?MoS'ct;Ss: ?^/S. 
 1853 - ^^"^*^"dge [Eng.] J. Deighton; jetc., etc.) ^ 
 
 vii, (li, 188 p. front, foI4 maps, plans. 22". 
 
 Restrictions on Use: 
 
 1. Il.ninil.al-Crossiiig of the Alps, B. c 218. 
 
 Library of Congress 
 
 DG247.2.E47 
 
 4-37256 
 
 TECHNICAL MICROFORM DATA 
 
 REDUCTION RATIO: /^v 
 
 FILM SIZE: Jj^^ 
 
 IMAGE PLACEMENT: lA Q^ IB IIB 
 
 DA IE FILMED: Jl-'_l7r_V.___ INITIALS ^^c^ 
 
 HLMEDBY: RESEARCH P UBLICATIONS. INC WOOnHRmnF FT/— 
 
>^^ >• MMMBMBiawiM*-* * " • • 
 
 BIBLIOGRAPHIC IRREGULARITIES 
 
 MAIN 
 EhfTRY: 
 
 -A 
 
 Mis, Rokj^rt . ^ 
 
 Bibliographic Irregularities in the O ri|^inal DocumPn> 
 List volumes and pages affected; include name of institution if filming borrowed text. 
 
 Page(s) missing/ not available: 
 
 yolumes(s) missing/not available:. 
 
 —jZ-IUegibl 
 
 e and/or damaged page(s): p . 
 
 Page(s) or volumes(s) misnumbered: 
 
 Bound out of sequence: 
 
 3^-5" 8^ 
 
 .Page(s) or illustration(s) filmed from copy borrowed from UttiV d^ 
 
 Other: 
 
FILMED IN WHOLE 
 
 OR PART FROM A 
 
 COPY BORROWED 
 
 FROM UNI V . OF 
 
 CALIFORNIA 
 
 BERKELY 
 
 9 
 

 1 
 
 
 
 r 
 
 Association for information and Image {Management 
 
 1100 Wayne Avenue, Suite 1100. 
 Silver Spring, Maryland 20910 
 
 301/587-8202 
 
 Centimeter 
 
 12 3 4 
 
 iiiiIiiiiIiiiiIiiiiIiiiiIiiiiIiiiiIh 
 
 II 
 
 ITT 
 
 Inches 
 
 I 
 
 1 
 
 5 6 7 8 
 
 iiiliiiiliiiiliiiiliiiiliiiiliiiiliii 
 
 * 
 
 1.0 
 
 I.I 
 
 1.25 
 
 llllillllllllllllllllll 
 
 10 
 
 L 
 
 TTT 
 
 11 12 13 
 
 liiiiliiiiliiiiliiiiliiii 
 
 T 
 
 T I I 
 
 i^ 2.8 
 
 2.5 
 
 y; 
 
 
 ta r-2 
 
 2.2 
 
 f^ 3,6 
 
 
 i:: ^ 
 
 ■ 80 
 
 2.0 
 
 i£ 
 
 
 •i u 
 
 tiititi. 
 
 
 1.8 
 
 1.4 
 
 1.6 
 
 14 15 mm 
 
 iiliiiiliiiii 
 
 MfiNUFRCTURED TO HUM STRNOnRDS 
 BY fiPPLIED IMAGE. INC. 
 

 ifS^Tf- ' .*"* 'JWri 
 
 
 
 
 
 -:;M;;'ai£-' 
 
 
 
 i. 
 
 « ♦■»■ 
 
 «'<f;< 
 
 •*?^ 
 
 ■%r 
 
 
 
 
 
 
 J.'w 
 
 ,!*«« « 
 
 J^^l(. I 
 
 ^i^'^ * r ' 
 
 -3f- 
 
 «T 
 
 
 •► ,ff'r», 
 
 *Vi r *♦■??. -'■ 
 
 
 
 
 »\^ 
 
 ?• t 
 
 ', * 
 
 ' r- 
 
 
 w*- 
 
 5 ,y^ §^ g* 
 
 rf r*.*- 
 
 ^' 
 
 tr > 
 
 » 
 
 F'- 
 
 
 ^ V 
 
 
 -»*»" 
 
 ..V,i^ 
 
 
 v' ^■'•:?'t'^ 
 
 C^:>^.. >'\V 
 
 .v*E'r:''"k.i 
 
 *^"3»''/ -'-'^ 
 
 
 
 -''.5v:....-'^ 
 
 
 liW' 
 
 i.-^*.)' 
 
 , i\ 
 
 H#^4?->: 
 
 
 ''i 
 
I -'-1 
 
 
 lu.. 
 
 
 H: 
 
 
 I 
 
 i 
 
 ! 
 
 ColumWa (Bnitiersftp 
 
 intljfCiipafllruigark 
 
 LIBRARY 
 
 :■::■[-: -m 
 
 i f 
 
 Ml 
 
 { 
 
 i .- 
 
 A -J 
 
TEEATISE 
 
 ON 
 
 V 
 
 '^n.xoiA. 
 
 X 
 
 * i > 
 
 Hih 
 
 It H 
 
 I! 
 
 hw 
 
 iil 
 
 HANNIBAL'S PASSAGE OF THE ALPS, 
 
 m WHICH HIS ROUTE IS TRACED OVER THE 
 
 
 LITTLE MONT C£NIS, 
 
 BY 
 
 ROBERT ELLIS, B.D., 
 
 FELLOW OF ST JOHN'S COLLEGE, CAMBEIDGE. 
 
 »I 
 
 ■J 
 
 CAMBRIDGE: J. DEIGHTON. 
 LONDON : JOHN W. PARKER AND SON. 
 
 1853. 
 
 I, 
 
OTambtftigc : 
 
 ^titttetJ at tlje mniijemts ^n^* 
 
 
 PREFACE. 
 
 IN the course of a series of excursions in the Alps, the 
 attention of the Author had naturally been drawn to 
 the subject of HannibaPs passage of those mountains. On 
 a subsequent examination of Polybius' account of that ex- 
 pedition, it appeared almost certain that the prominent 
 and distinguishing characteristics of the pass which Han- 
 nibal crossed could only be found on the pass of the Mont 
 ) Cenis. A general acquaintance with the passes, by which 
 it had been conjectured that Hannibal entered Italy, led 
 at first to this conclusion. The close proximity to the 
 plains of Piedmont of the summit of the pass which Han- 
 nibal crossed, the precipitous nature of the descent into 
 Italy, and the immense losses which the Carthaginians 
 incurred in consequence, all tended to identify this pass 
 with that of the Mont Cenis. It was also beyond a doubt, 
 that from neither the Great nor the Little St Bernard, nor 
 from the Mont Genevre, was it possible to. see the Italian 
 plain. None of the roads over these passes, nor any 
 accessible point in their neighbourhood, could command 
 the view which the Carthaginian army saw from the crest 
 of the Alps. The situation of the Mont Cenis, as might 
 be judged from the map, rendered it, however, not impro- 
 bable, that such a view might be obtained from the imme- 
 diate neighbourhood of the road over its summit. As 
 this route, then, had a considerable amount of a priori 
 \ probability in its favour, a probability shared by no other 
 pass, it seemed likely that an examination of it in detail 
 might lead to its complete identification with the line 
 of Hannibal's march into Italy. Accordingly, in the month 
 of July, 1852, and in the months of April and May, 1853, 
 
 I 
 
 ) hi 
 
 I 
 
 /' 
 
 \i 
 
 Q '"' R 2 
 
% 
 I 
 
 J 
 
 iv 
 
 PREFACE. 
 
 the requisite examination was made. A period of nearly 
 six weeks was, altogether, spent upon the investigation, 
 some days being lost in consequence of unfavourable 
 weather. A great quantity of snow was found upon the 
 ground during the spring months, and rendered the cha- 
 racter of the pass more in accordance with what it would 
 have been at the time of Hannibal's passage of the Alps. 
 
 It is now, it may be said, universally admitted, that the 
 great test, by which the claims of every pass must be tried, 
 is the narrative of Polybius. The following investigation 
 has, accordingly, been conducted on this principle. In the 
 translation of Polybius, which accompanies it, the exact 
 meaning of the historian, and his peculiarities of narration, 
 have been preserved as far as possible. To gain this end, 
 of the utmost importance in the present case, nothing 
 beyond a merely literal version of the Greek has been 
 attempted. 
 
 Less attention has been paid to Livy'^s narrative ; but 
 his evidence has not been neglected. An examination of 
 the objections brought against the Mont Cenis, and an 
 enquiry into the antiquity of that pass, conclude the inves- 
 tigation. The testimony of Polybius, and that of Livy, 
 have been kept, almost entirely, distinct from one another; 
 and the question of the old roads across the Alps has also 
 been considered by itself, and a third and independent 
 body of evidence deduced from its discussion. 
 
 The thanks of the Author are due to the Syndics of 
 .the Pitt Press for the liberality with which they have 
 undertaken the printing of the present work. 
 
 St John's College, 
 
 Dec. 23, 1853. 
 
 nm 
 
 i 
 
 CONTENTS. 
 
 CHAPTER I. 
 
 Introduction. — Brief notice of Hannibal's march from New Car- 
 thage. — Polybius* estimation of the lengths of the different 
 parts of the whole march from New Carthage to the plains of 
 
 Italy. A peculiarity in Polybius' mode of narration observed. 
 
 Division of the march from the passage of the Rhone to 
 
 the country of the Insubrian Gauls into seven stages 
 
 CHAPTER II. 
 
 Polybius' narrative of Hannibal's march, from the passage of the 
 Rhone to the capture of Turin 
 
 CHAPTER III. 
 
 Discussion of the narrative of Polybius, from the passage of the 
 Rhone to the commencement of the ascent of the Alps. — Iden- 
 tification of the Scaras of Polybius with the modem Ishre.-- 
 The Island of Polybius the district between the Rhone and 
 the Isere. — Polybius only partially acquainted with this district. 
 — Conclusion drawn that Hannibal must have marched up the 
 left bank of the Isere, and not up the Rhone, from the junc- 
 tion of the two rivers • • 
 
 CHAPTER IV. 
 
 Discussion of the narrative of Polybius, from the commencement 
 of the ascent of the Alps to the arrival at the summit of the 
 pass. — The defile at the commencement of the ascent. — Con- 
 clusions drawn as to its nature and the character of the neigh- 
 bouring country. — The * strong white Kock.' — Features of the 
 country in its vicinity. — The* rock' at a distance of many miles 
 from the summit of the pass ....•• 
 
 PAGE 
 
 8 
 
 16 
 
 V 
 
 29 
 
VI 
 
 CONTENTS. 
 
 CHAPTER V. 
 
 i 
 
 \ I' 
 
 I 
 
 PAGE 
 
 Discussion of Polybius' narrative of the events which took place 
 at the summit of the pass and on the descent to Turin. — The 
 plains of Italy visible from the summit of the pass. — Precipi- 
 tous nature of the descent. — The path destroyed by a landslip. 
 — Manner in which the path was re-constructed. — Hannibal 
 arrives in the country of the Taurini. — Journal of the march 
 from the confluence of the Rhone and Isere to the commence- 
 ment of the plains of Italy. — Conditions for the determination 
 of the pass which Hannibal crossed . . ... 60 
 
 CHAPTER VI. 
 
 Improbability of HannibaVs having crossed the Alps by the Great 
 or Little St Bernard, or by the Mont Genevre. — Probability of 
 his having crossed the Mont Cenis. — Determination of the point 
 where Hannibal would have left the Isere. — The oldest line of 
 road over the Mont Cenis traced and described. — The Little 
 Mont Cenis selected as the pass which Hannibal crossed, in 
 preference to the Great Mont Cenis . . . . .68 
 
 CHAPTER VII. 
 
 Application of the conditions deduced from Polybius* narrative to 
 the route of the Mont Cenis. — Accordance of the actual dis- 
 tances with those given by Polybius. — Identification of the 
 defile at the commencement of the ascent of the Alps with the 
 gorge of Le Fay, — Probable identity of the town of the Allo- 
 broges and the modern Allevard, — Identification of the "strong 
 white rock" with the rock of Baune. — Arrival at the Col of the 
 Little Mont Cenis ........ 
 
 89 
 
 )\ 
 
 CHAPTER VIII. 
 
 Application of the conditions deduced from Polybius* narrative to 
 the route of the Mont Cenis concluded. — Encampment of Han- 
 nibal on the Plateau of the Little Mont Cenis. — Point from 
 which the plains of Italy were most probably seen. — Precipitous 
 character of the descent from the Mont Cenis into Italy. — Pro- 
 bable situation of the place where the path was destroyed. — 
 Arrival at the commencement of the plains of Italy. — Siege 
 and capture of Turin ....... 
 
 114 
 
 I 
 
 wmm 
 
 
 / 
 
 CONTENTS. 
 
 vii 
 
 CHAPTER IX. 
 
 LivVs narrative of Hannibars passage of the Alps.-Its character^ 
 -Use to be made of it.-Hannibars line of approach to the 
 Alps as indicated by Livy.-His account of the events which 
 fook^Tace on the summit of the pass.-^The broken path.- 
 took place —Arrival of Hannibal m the 
 
 Singular story connected witn it. Arnvai 
 
 country of the Taurini .•••••* 
 
 CHAPTER X. 
 
 Examination of objections brought against the Mont Cenis . . 
 
 PAGE 
 
 129 
 
 147 
 
 :U' 
 
 CHAPTER XI. 
 
 . ,, w^„x ppnis — First mentioned by name 
 Antiquity of the pass of *« M"^ Cems. * ^^^^ ^^^^^^ _ 
 
 in the year 755 A d.— Probably "^n"^" century.-The 
 
 Described by An.mia.us ^l^^'"^-^:}^ ^1^:^^-^^^-^-- 
 road over the Little Mont Cen.s laid d";" >° 'J _Probably 
 
 table.-The Mont Cenis '^^<^^\^\^^^Z c^n^ of tU Tal 
 mentioned by Strabo as tlu> passthroug^^ tU country J ^ ^^^ 
 
 rini, which Hannibal cro8»«Z.— Conclusion . 
 
 APPENDIX. 
 
 A. fv^m Ttalv bv the Little St Bernard, 
 
 A. Note on the Roman roads fiom Italy, oy v ^ ^ ^^^ 
 
 to Vienne and Geneva ' , e a^^Lmaam . ■ ^^T 
 
 B. Note on the position of Ocduu. and of Sc^ngon,^ 
 
 H 
 
 i 
 
 
\ 
 
 A TREATISE 
 
 ON 
 
 1 
 
 Directions for the insertion of the Maps and Plans. 
 
 Map of the route of the Mont Cenis . at the beginning of tho book. 
 Map of the Roman roads ... at the end of the book. 
 
 Plan of the gorge of Le Fay .... to face p. 89 
 Country in the neighbourhood of the Rock of Baune . . . 112 
 Mountain-sido on the descent from the Mont Cenis . . . 128 
 Part of the Peutingerian table . . . .... 159 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 CHAPTER I. 
 
 l.mRAK^ 
 
 Introduction.— Brief notice of Hannibal's march from New Carthage.— H . y/ 
 Polybius' estimation of the lengths of the different parts of the — ^ 
 whole march from New Carthage to the plains of Italy. — A pe- 
 culiarity in Polybius* mode of narration observed. — Division of the 
 march from the passage of the Rhone to the country of the Insubrian 
 Gauls into seven stages. 
 
 A PERIOD of more than twenty years of peace suc- 
 ceeded the close of the first Punic war. It was not 
 however a peace which could be expected to endure. 
 The^ mutual exhaustion of the contending parties had 
 obliged them for a time to desist from the further prose- 
 cution of hostilities : but the power of neither the Romans 
 nor the Carthaginians was inwardly shattered; and the 
 rivalry and hatred with which they had long regarded each 
 other were necessarily increased rather than diminished 
 by the losses which both had suffered. These losses, 
 , severe as they had been, were gradually repaired ; each 
 * nation recovered its former vigour, and looked forward to 
 the renewal of a strife, which the deeply-rooted animosity, 
 the clashing interests, and the ambitious spirit of both 
 parties, plainly forewarned them could have no end but in 
 the complete subjection of Rome or Carthage. The first 
 Punic war had established an inveterate rivalry between 
 the two cities : the event of the second war was to deter- 
 mine which should fall. 
 
 
 t,^--^*. 
 
u>' 
 
 2 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 M 
 
 \i 
 
 
 The struggle which was to decide this question at 
 length approached. In the year 219 b. c. the peace be- 
 tween the Romans and Carthaginians was forraally broken; 
 hostile declarations were exchanged at Carthage, and pre- 
 parations for the contest were made on both sides. The 
 news of the rupture reached Hannibal at New Carthage in 
 Spain, a country which he had in a great measure reduced 
 imder the Carthaginian yoke; and where the continual 
 growth and consolidation of the power of their enemy had 
 given much umbrage and disquiet to the Romans. It was 
 there he meditated the invasion of Italy and the fall of 
 Rome, and resolved upon that march which has associated 
 his name for ever with the Alps. His preparations for the 
 invasion had commenced before he was aware of the 
 rupture of the peace, and by the spring of the year 218 
 they were finally completed. 
 
 A fortunate conjuncture of circumstances had enabled 
 him to secure allies in Italy. The Romans were at this 
 time, and had been for some years previously, making gra- 
 dual encroachments upon the territories of the Cisalpine 
 Gauls : and the Gauls, feeling that they were of themselves 
 incapable of resisting the growing might of Rome, readily 
 embraced the opportunity of forming an alliance with the 
 great enemy of the Roman name. In anticipation of such 
 an event, and before he had received the news that the 
 Romans had declared war against the Carthaginians, Han- 
 nibal had despatched an embassy from Spain into Cisalpine 
 Gaul, to ascertain and confirm the disposition of the inhabit- 
 ants towards the Carthaginian alliance, and to discover also, 
 whether the passage of the Alps was practicable or not. 
 
 The return of his envoys he was now awaiting. At 
 length they arrived at New Carthage, and brought back 
 the intelligence, that the Cisalpine Gauls were very favour- 
 ably disposed towards the Carthaginians. The passage 
 of the Alps they also reported to be practicable, although 
 exceedingly difficult and unavoidably attended with great 
 labour \ Impossible, however, they were assured that it 
 was not. For the Gauls themselves had frequently, not 
 
 1 Polvb. m. 34. 
 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 3 
 
 only in remote, but also in recent times, crossed the 
 mountains with large armies': and although the march of 
 the Carthaginian host, encumbered with its elephants and 
 baggage, would be much less easily performed than that of 
 an army of barbarians ; still the difference was not so great 
 as to lead them to conclude the passage impracticable. 
 The receipt of this intelligence determined the course of 
 Hannibal. His resolution of penetrating across the Alps 
 into Cisalpine Gaul was finally taken, and at the end of 
 May the army began its march from New Carthage. 
 
 It consisted at first of 90,000 foot and 12,000 horse, 
 besides elephants and baggage-animals. All, however, did 
 not quit the Spanish territory. Some were lost in sub- 
 duing the nations between the Ebro and the Pyrenees: 
 and when the march was resumed after the reduction of 
 these tribes, a detachment of 11,000 was left behind under 
 Hanno, and an equal number of men were dismissed to 
 their own homes. 50,000 foot and 9000 horse were all 
 that crossed the Pyrenees and entered Gaul. After the 
 passage of the Rhone this number was diminished to 38,000 
 foot and 8000 horse 2. Of these, 20,000 foot and 6000 
 horse alone survived the passage of the Alps, and suc- 
 ceeded in reaching the plains of the Po. 
 
 HannibaPs route from New Carthage lay, as just inti- 
 mated, northward; and was directed along the coast of 
 Spain. The river Ebro was crossed, the nations between 
 it and the Eastern Pyrenees were subdued, the extremity 
 of those mountains was passed, and Transalpine Gaul en- 
 tered. After the passage of the Pyrenees, Hannibal still 
 kept for a time near the sea, but at length turned inland, 
 and crossed the Rhone at a distance of nearly four days' 
 march from the sea. Proceeding subsequently along the 
 left or eastern bank of the Rhone for four days more, he 
 
 1 Polyb. ni. 48. 
 
 2 It is difficult to account for the loss of 13,000 men between the 
 Pyrenees and the Rhone. The ordinary casualties seem insufficient to 
 explain it. Perhaps a number of the Iberians had deserted on enter- 
 ing Gaul, and returned to their own country, having been disappointed 
 at not being included in the number of those whom Hannibal had allowed 
 to do 60. 
 
 1—2 
 
 /,■ 
 
 If 
 
 \J 
 
MM 
 
 ^ 
 
 4 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 reached, at the end of that period, a district called the 
 Island, of which Polybius gives a particular description. 
 Having left the Island, Hannibal struck off towards the 
 Alps, effected his passage across them, and descended into 
 the plains of the Po. The whole of this march, from New 
 Carthage to the Italian plains, has been divided by Poly- 
 bius (iii. 39) into five stages, which terminate, respectively, 
 at the passage of the Ebro, at Ampurias (Emporium), at 
 the passage of the Rhone, at the foot of the first Alpine 
 ascent, and at the commencement of the plains of Italy. 
 The last two of these stages, in which the passage of the 
 Alps is included, are all with which this book is immedi- 
 ately concerned. 
 
 Before, however, proceeding to the examination of this 
 part of Polybius' narrative, it will be necessary to give 
 the distances comprised in each stage of the march, from 
 New Carthage to the commencement of the plains of Italy, 
 They are laid down by Polybius as follows (iii. 39) : 
 
 " From this city (New Carthage) to the river Ebro the 
 distance is 2600 stadia : again from this river to Emporium 
 1600 stadia ; and from thence also to the passage of the 
 Hhone about 1600 stadia : for these distances have been 
 now measured in paces and marked carefully by the Ro- 
 mans at intervals of eight stadia. From the passage of 
 the Rhone, as they (the Carthaginians) marched along the 
 bank of the river, in the direction of its sources, as far as 
 the ascent of the Alps on the way to Italy, there was a 
 further distance of 1400 stadia. There still 7*emains to be 
 estimated the passage of the Alps, a march of about 1200 
 stadia; and, after crossing these mountains Hannibal 
 would immediately arrive at the plains of Italy lying around 
 the Po. The whole length of his march from New Car- 
 thage would consequently amount to about 9000 stadia ^" 
 
 The first thing requisite to be observed here is Poly- 
 bius' scale of measurement. He computes, it will be per- 
 ceived, by lengths of 200 stadia, or 25 Roman miles. 
 Doubtless, with his deficient means of estimating distances. 
 
 )i' 
 
 ' mi'\ 
 
 1' 
 
 1 The sum of the distances given by Polybius only amounts really to 
 8400 stadia. 
 
 HANNIBAL'S PASSAGE OF TfiE ALPS. 5 
 
 he found himself unable to determine the lengths of the seve- 
 ral marches more closely, at least in such parts of the route 
 as the Romans had not measured. Yet for these parts also 
 he employs the same rule of computation, it not being ap- 
 parently his object to aim at greater accuracy. This fact 
 must always be borne in mind in the investigation of Han- 
 nibal's route, in order that too much stress may not be 
 laid upon discrepancies of distance, when not great : while, 
 at the same time, due attention should still be paid to the 
 statements of distances, given by so careful an authority as 
 Polybius, and a too great laxity avoided in accommo- 
 dating them to actual routes. It must therefore be remem- 
 bered that Polybius' measure of distance is 25 Roman 
 miles, in computing the length of a march of several days' 
 duration. 
 
 The passage marked in italics, in the quotation given 
 above from Polybius, is more particularly connected with 
 the following investigation. It supplies at once one of the 
 conditions, which the pass to be identified with that which 
 Hannibal crossed must satisfy. This condition may be 
 
 thus stated: 
 
 *' The length of the route over the Alps, beginning at 
 the commencement of the ascent of the mountains, and 
 terminating at the commencement of the plains of Italy, 
 must be about 150 Roman miles ^" 
 
 That the termination of this part of the march was at 
 the commencement of the plains of Italy, is clearly stated by 
 Polybius. That " the ascent of the Alps on the way to Italy" 
 means the place where the route first became mountainous, 
 the point where the army was first obliged to ascend the 
 mountains, — this seems also sufficiently plain ; for it ap- 
 pears scarcely probable that any other meaning can be 
 attached to the expression of Polybius, ti^v dva(io\t]u 
 rwv "AXTreo)*' t»;V ek 'IraXiai^. There is however a certain 
 amount of vagueness in it, and a different interpretation 
 might be possible. 
 
 The narrative of Polybius, from the time when the 
 march was resumed after the passage of the Rhone, up to 
 
 
 1 Condition II. 
 
9 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 )) 
 
 f ^ 
 
 the time of Hannibal's arrival in the plains of Italy, will 
 now be given in full, with the exception of three chapters 
 and part of a fourth. The passages omitted include the 
 end of the fifty-sixth chapter, and all the fifty-seventh, 
 fifty-eighth, and fifty-ninth chapters. These parts of the 
 narrative are passed over, as not bearing upon the ques- 
 tion of Hannibal's passage of the Alps, but as relating 
 either to the operations of the Eoman armies, or else 
 containing general reflections upon the subject of history. 
 
 There is, however, one peculiarity in Polybius ' style 
 of narration, especially in this part of his history, upon 
 which it will previously be necessary to make some observa- 
 tions. The peculiarity alluded to is this : that the histo- 
 rian, before entering into the details of a particular march, 
 event, or military transaction, gives, in a few lines, what 
 may be regarded as a short statement or summary of the 
 occurrences which took place at that particular period. 
 Having done this, he proceeds to make such observations, 
 and give such explanations, as appear necessary, or to 
 narrate at length the various circumstances that attended 
 the facts in question, whenever they were of such import- 
 ance as to deserve minute consideration. The short sum- 
 mary serves frequently, in point of fact, as an argument to 
 the succeeding and more detailed account. The whole of 
 that portion of Polybius' narrative now about to be given 
 may be considered with much probability as written in this 
 manner, and will divide itself into seven parts. Four of 
 these relate to the actual passage of the Alps ; two to 
 the march from the place where the Rhone was crossed, to 
 the commencement of the mountains ; and one to the 
 march from the foot of the Alps, on the Italian side, to 
 the territories of the Insubrian Gauls, the allies of Han- 
 nibal. The first division will consist of the march from 
 the passage of the Ehone to the district called the Island • 
 the second, the march from the Island to a certain defile 
 and town at the commencement of the Alps ; the third 
 the march from the town just mentioned to the nei^h^ 
 bourhood of a certain X.vK^Trerpou 6xvp6^, or ' strong i 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 7 
 
 white rock,' where the army encountered great danger from 
 an attack of the Alpine Gauls ; the fourth, the march 
 from this rock to the summit of the pass; the fifth, the 
 circumstances which took place while the army remained 
 on the summit of the pass ; the sixth, the descent from the 
 summit of the Alps to the commencement of the plains of 
 Italy ; and the seventh, (all of which will not be given) the 
 march from the foot of the Alps to the country of the 
 Insubrians. 
 
 These form the seven parts, into which the narrative 
 seems to be divided. The correctness of the supposition, 
 that this mode of narration was adopted by Polybius, 
 will be best seen by an inspection of the historian's own 
 words. Accordingly, those parts of his account, which are 
 supposed to be the summaries in question, will, in the 
 extract from Polybius which now follows, be distinguished 
 by being printed in capital letters. Those passages also of 
 the narrative, which bear more particularly on the present 
 question, and afford a clue to the determination of the 
 features of the country through which Hannibal passed, will 
 be brought prominently forward by being printed in italics. 
 Polybius' account of the march, thus arranged, will form 
 the substance of the next chapter. 
 
 ( 
 
 / 
 
 iff 
 
 1 * 
 
 Strong,' of courgo, in a military point of vi 
 
 icw. 
 
 ^ 
 
II 
 
 f I 
 
 CHAPTER II. 
 
 Polybius' Narrative of Hannibal's March, from the passage of the 
 
 Rhone to the capture of Turin. 
 
 Polybius. Book in. Part of Chap. 49. 
 
 HANNIBAL, HAVING continued his march for four successive 
 DAYS FROM THE PASSAGE (OF THE RhONE) CAME TO THE DISTRICT 
 CALLED THE ISLAND, A POPULOUS AND CORN-PRODUCING TRACT OF COUN- 
 TRY, WHICH DERIVES ITS NAMK FROM THE VERY CIRCUMSTANCE (OF ITS 
 
 formation). For the Rhone flowing on one side of this district, and 
 a river called the Scaras^ on the other, sharpen it into a point at 
 their confluence. It is similar in size and form to the so-called 
 Delta in Egypt ; excepting that, in the case of the Delta, the sea 
 forms one side (i. e. the third side) by joining the mouths of the 
 rivers (which form the other two sides) ; whereas the third side 
 of the Island is formed by mountains difficult of approach and 
 entrance, and, it may he said, almost inaccessible. When Hannibal 
 arrived at the Island, he found two brothers contending for the 
 royal power, and encamped with their armies opposite each 
 other ; and when the elder brother tried to induce him to second 
 his efforts in acquiring the sovereignty, he listened to the over- 
 tures which were made to him : for it was sufficiently evident 
 what advantages he would derive from the assistance of the Gaul 
 at the present juncture. By their combined forces, the second 
 claimant of the sovereignty was attacked and expelled ; and Han- 
 nibal afterwards received great assistance from the victor. For 
 he not only abundantly supplied the army with corn and all other 
 necessaries, but also exchanged such of their arms and accoutre- 
 ments as were old and worn out, and equipped afresh all the 
 forces at a time when such aid was very opportune. He also 
 furnished the mass of the soldiers with clothing and shoes, a pro- 
 vision which was of great service for the passage of the mountains. 
 The most signal benefit however which he conferred upon them 
 was, that as they were apprehensive of danger while they marched 
 through the territory of the Gauls who are called AUobroges, he 
 covered their rear with his own troops, and secured for them a 
 safe advance until they approached the passage of the Alps. 
 
 1 The manuscripts have ^Kapas or ^Kwpas. 'Icrapar is only conjectural. 
 It has been thought better to adhere to the MS. reading. 
 
 f>CT 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 9 
 
 Chap. 50. 
 Hannibal, in the course of ten days, having advanced along 
 
 THE river-side FOR A DISTANCE OF 800 STADIA, COMMENCED THE 
 ASCENT OF THE AlPS, AND FOUND HIMSELF IN A SITUATION OF THE 
 
 GREATEST DANGER. For as long as the army continued in the 
 plain, all the petty chieftains of the AUobroges refrained from 
 attacking them, partly from the fear of their cavalry, and partly 
 from fear of the Barbarians who formed their escort. But when 
 these last had returned to their own country, and Hannibal, with 
 his army, was advancing towards the places ivhich were difficult of 
 passage, the chieftains of the AUobroges coUected a sufficient body 
 of forces, and took possession of the advantageous posts along the 
 road by which Hannibal was obUged to make his ascent. Had 
 they concealed their plans, they must have altogether destroyed 
 the Carthaginian army; and, even after their purpose was de- 
 tected, they inflicted great injury upon the troops of Hannibal, 
 although the loss which they themselves incurred was equally 
 great, ^'por, when the Carthaginian general discovered that they 
 had occupied the advantageous posts, he halted his troops, and 
 encamped near the heights : he then sent forward a party of the 
 Gauls, who acted as his guides, to discover the plan of the enemy, 
 and the whole scheme of their operations. The Gauls executed 
 his commission ; and the Carthaginian general found, from the 
 intelligence they brought, that the enemy guarded very vigUantly 
 their posts during the daytime, but withdrew at night to a town 
 in the neighbourhood. To counteract therefore their designs, he 
 determined upon operations of the foUowing nature. He set his 
 army in motion and advanced in open view ; and, when he drew 
 near to the difficult places, he encamped at a short distance from 
 the enemy. When night came on, he ordered the watch-fires to be 
 lighted, and left in the camp the mass of his forces ; while he 
 himself, with the most active of his troops, purposely equipped for 
 such a service, penetrated in the night through the defile, and seized 
 the posts which had been before occupied by the enemy, who had now 
 retired to the town, according to their usual practice. 
 
 Chap. 51. 
 
 When this had been accompUshed, and day appeared, the 
 Barbarians, observing what had taken place, desisted for a 
 time from their original design; but when they subsequently 
 saw the numbers of the baggage-animals and the cavalry 
 laboriously, and, in an extended line, defiling through the diffixiult 
 places, they were induced, by this circumstance, to attack the 
 line of march. When this was done, and the Barbarians had 
 
 V. 
 
 i 
 
■n 
 
 \ 
 
 i 
 
 10 
 
 HANNIBAL'S PASSAGE OP THE ALPS. 
 
 fallen upon them on many points, a great loss ensued to the Car- 
 thagmians especiaUy in horses and baggage-animals : and this 
 
 !/^//,rr ' z"" V '?r'' ^^ 'l' "'"'''^•'"fi' *"*'»y' ^ ^y *^ mature 
 of m ground, tor the way by which the Carthaginians had to 
 
 advance was not only narrotv and rugged, but also precipitous, so 
 
 that at every shock or commotion, numbers of the animals with their 
 
 burdens ivere carried doion the precipices. This confusion was 
 
 mamly caused by the horses which were wounded. Some of 
 
 these, rendered unruly by their wounds, fell back upon the ba- 
 
 gage-animals ; and others, in their impetuous advance drove 
 
 before them all that fell in their way upon this difficult ground 
 
 l7rfr f '^ri?'?" ''"' ^'"'' "*'*'^^- W'^*^" Hannibal saw this, 
 and reflected that even those who escaped such dangers would 
 
 wero "''.^■'l""''*^ °.fP'-««?"ation ultimately, if the baggage-animals 
 
 weie lost, he set m motion the soldiers who, during the^ight, had 
 
 taken possession of the heights, and rushed down to the'succo^r 
 
 ot the foremost column in the line of march'. By this move- 
 
 ment many of the enemy were destroyed, in cLeauen^ of 
 
 Hanmbal attacUng them from the heights above them ZZ lot 
 
 Tumult :„rr? '" f """i "™^ "«* '^-^"""^ ^'•-'- ^^i z 
 
 tumult on the line of march was increased on both sides fAllc 
 broges and Carthaginians) in consequence of the shoS of he 
 
 BuT wZ"h TX ""f. '''''' intermingling in the confl ct 
 But when Hanmbal had destroyed the greater number of the 
 Allobrogcs and forced the remnant of them to flrhomewards 
 then al the baggage-animals and horses which Jcrc kft with 
 much labour and trouble effected their passage throu-^h the difficult 
 
 h Tould areS'^' ""r ■'"' ^'^"^^'^' «^ --y of lis LmS 
 he could after this perilous service, and fell upon the town from 
 
 which the enemy had made their attack : and finding halmos" 
 deserted, m consequence of all its inhabitants having |o„e outTn 
 hope of plunder, he became master of it. This capture ensured 
 him many present and subsequent advantages. mZovered^m 
 mecUaU possession of a number of his horses^and be^stZZurZ' 
 and the men who had been made prisoners with them, and gaS a 
 supply of corn and cattle sufficient for two or thrPP IvT!^ 
 sumption. But the chief advantage whicrre^ulted was {LtTe" 
 inspired such great alarm among the people adiar^nt fr? f^ 
 
 r-iJ 'hi^ """^ "^ ''''- '- ''- futL^tntSiiUVt: 
 
 » Read here with Casaubon, Gronoyius. and Bokker, npoKa^oiai 
 
 whorithc^Snttth *'" '''f™'7''«- ho then was. w'hich was that 
 wuere tno ascent to the mountains first bo<van. 
 
 HAISNIBAL'S PASSAGE OF THE ALPS. 
 
 Chap. 52. 
 
 11 
 
 Having then encamped there and remained one day, he pro- 
 ceeded onwards. During the days immediately roLLOwiNG he 
 
 LED ON THE ARMY SAFELY FOR SOME DISTANCE : BUT ON THE FOURTH 
 DAY HE AGAIN HAD TO ENCOUNTER GREAT DANGERS. For the people 
 
 who lived along his route, having formed a treacherous design 
 airainst him, went out to meet him with crowns and branches of 
 trees , — now these among Barbarians are considered almost uni- 
 versally as tokens of amicable intentions, as is the case with the 
 herald's wand among the Greeks. But Hannibal, cautious of 
 giving too easy credence to them, made careful enquiry into 
 their aims and intentions. They alleged their perfect knowledge 
 of the capture of the town, and the destruction of the peoi)le 
 who had attempted to injure Hannibal ; and affirmed that those 
 events had induced them to seek that interview, as they were 
 anxious to prevent the occurrence of mutual hostilities. They 
 also promised to give some of their own people as hostages to 
 the Carthaginians. Hannibal continued for a long time on his 
 guard against them, and distrusted their representations; but 
 when he considered, that if he were to accept their overtures, 
 he would, in all probability, render them more cautious^ and well 
 disposed ; whereas, if he rejected them, he would certainly incur 
 their open hostility; — (for these reasons) he assented to their 
 proposals, and affected to engage in alliance with them. The 
 Barbarians gave the hostages they had promised, supplied the 
 troops with cattle in great abundance, and so unreservedly 
 placed themselves in the hands of the Carthaginians, that the 
 latter trusted to them as guides through the difficult places they 
 next encountered. Wlien these Barbarians had led the way in the 
 van for two days, the tribes before mentioned having collected 
 their forces and followed on the rear of the Carthaginians, at- 
 tacked them as they were passing through a difficult and precipi- 
 tous ravine. 
 
 Chap. 53. 
 
 All would again have been lost, had it not been that Hannibal, 
 still to a certain extent distrusting the Barbarians^ and providing 
 against the issue, had placed his baggage-animals and cavalry in 
 the van, and his heavy infantry in the rear. The protection 
 which these afforded mitigated the calamity, for the heavy infan- 
 try succeeded in withstanding the attacks of the Barbarians, Not- 
 
 1 This word probably relates to the safety of the hostages, which 
 would be compromised by the bad faith of the Barbarians. 
 
 i 
 
mneSBMa 
 
 ». ..— - ■ ■ 
 
 ^SiT' 
 
 m 
 
 12 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 withstanding this circumstance however, a considerable number of 
 men, baggage-animalsy and horses were destroyed. For the Bar- 
 harians, being in possession of the heights, and marching along the 
 mountain-sides in a parallel line to the Carthaginian columns, kept 
 rolling doivn rochsfrom above upon part of the army, and wounded 
 others from a nearer distance by hurling stones at them. So utter 
 was the consternation, and so great the danger occasioned by 
 this attack, that Hannibal was obliged to pass the night with half 
 his army near a certain strong white roch, away from the cavalry 
 and baggage-animals, thereby affording protection to the latter, 
 until they had with difficulty defiled through the ravine, an 
 operation which occupied the whole night. Hannibal hav- 
 ing ON THE FOLLOWING DAY, WHEN THE ENEMY HAD RETIRED, RE- 
 JOINED HIS CAVALRY AND BAGGAGE-ANIMALS, LED ON HIS ARMY TO 
 THE HIGHEST SUMMITS OF THE AlPS, WITHOUT MEETING AGAIN WITH 
 ANY CONSIDERABLE BODY OF THE BARBARIANS, ALTHOUGH PARTS OF HIS 
 ARMY WERE HARASSED BY THEM AT VARIOUS PLACES ON THE ROAD. 
 
 For they, watching their opportunity, assaulted and carried off 
 the baggage-animals, sometimes from the rear, and sometimes 
 from the van, of the line of march. Upon these occasions, the 
 elephants were of very great service ; for the Barbarians were so 
 much alarmed at the extraordinary appearance of these animals, 
 that they were deterred from attacking any part of the line of 
 march where the elephants were to be found. On the nIx^th day 
 Hannibal arrived at the summit of the mountains ; and, en- 
 camping THERE, REMAINED TWO DAYS, AS HE WISHED TO GIVE SOME 
 repose TO THOSE TROOPS WHO HAD ALREADY ARRIVED SAFELY, AND 
 
 TO WAIT FOR THOSE WHO HAD FALLEN BEHIND. During this period 
 many of the horses which had broken loose in their fright, and 
 many of the baggage-animals which had got rid of their burdens, 
 unexpectedly joined them in the camp, having followed appa- 
 rently the tracks of the army. 
 
 Chap. 54. 
 
 The snow was now gathering around the summits of the moun- 
 tains (for it was near the time of the setting of the Pleiades ^) ; 
 and Hannibal, observing that the soldiers were much dispirited 
 by the sufferings they had already undergone, and the anticipa- 
 tion of those which yet awaited them, called them together and 
 endeavoured to raise their spirits. The prospect of Italy offered 
 itself as the only means by which this result could be attained. For 
 Italy lies extended beneath these mountains in such a manner, 
 that, when both are seen together, the Alps seem to occupy the 
 
 1 The last week in October. 
 
 '/ 
 
 / 
 
 / 
 
 H 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 13 
 
 position of a citadel to the whole of Italy. Accordingly Hanni- 
 bal, pointing out to his men the plains circumjacent to the Po, and 
 reminding them of the goodwill which the Gauls who dwelt there 
 entertained towards them, and at the same time indicating to 
 them the quarter in which Rome itself lay, succeeded to a certain 
 degree in raising their drooping spirits. The following day he 
 
 BROKE up HIS CAMP, AND COMMENCED HIS DESCENT, DURING WHICH HE 
 ONLY SUFFERED FROM FURTIVE ATTACKS ON THE PART OF THE ENEMY ; 
 YET IN CONSEQUENCE OF THE SNOW AND THE NATURE OF THE GROUND, 
 HIS LOSS WAS NEARLY AS GREAT AS IT HAD BEEN DURING THE AS- 
 CENT. For as the way down was narrow and steep, and the snow 
 made every one uncertain of his path ; all, both men and animals, 
 that fell out of the track or missed their footing, were carried down 
 the precipices. Still, however, they bore these hardships with 
 the patience of men already habituated to similar calamities. 
 But when they arrived at a place where it was impossible for the 
 elephants or baggage-animals to pass on account of the narrowness 
 of the way {the broken ground which previously occurred here for 
 a length of a stadium and a half having recently suffered additional 
 disruption), the army was again greatly dispirited and perplexed. 
 The Carthaginian general at first proposed to make a circuit 
 round the difficult ground, but as the fresh snow that had fallen 
 rendered this impracticable, he desisted from his attempt. 
 
 Chap. 55. 
 
 For the circumstances were pecuHar and extraordinary. 
 Upon the snow previously existing, and which had remained 
 from the preceding winter, the snow of the present year had 
 lately fallen. Now they found this latter snow easy to penetrate 
 and soft, because it was fresh, and not as yet of any depth. But 
 as soon as they had trodden through this, and reached the com- 
 pacted snow beneath, they were unable to penetrate it; but 
 both their feet sUpped away along its surface, as is sometimes the 
 case with people walking over ground of which the surface is 
 covered with mud. But what ensued was still more calamitous. 
 As the men were unable to penetrate the lower snow, whenever 
 upon falling they tried to support themselves upon their hands and 
 knees and to get up again, they only found that they made them- 
 selves slip still more, and carried with theift whatever they laid 
 hold upon for support — for the declivity was one of excessive 
 steepness. The baggage-animals, on the contrary, when they fell, 
 broke through the surface of the lower snow in their attempt to 
 rise, and there remained firmly fixed with their burdens, in con- 
 sequence of the weight they bore, and the compact nature of 
 the old snow. Accordingly Hannibal, giving up all hope of car- 
 
 
IIUJILUJII.IBJ 
 
 14 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 rying out his first plan, encamped about the ere^^f nf n,. ■ i 
 having first cleared away the snow upon it ZLf ?'^^'\ 
 large numbers of his soldiers i„hZZ^ ., *'*^" employed 
 
 of great labour and dTfficuUv ^ ^*"^,"^ '^'« P'-^^P^e, a task 
 
 Hannibal, therefore^ male /em pass Tonee andThreri" '^^ 
 loi:"t: Str^e ^" He Z '°""', "^'-^^ ^^'erthTm 
 niidians in tur^n^ u^o^^he ton^t^ e^Tof t^^^^^^^^^^^^^ 
 
 Chap. 56. 
 
 .ese^rc^Jt:^.r Xr^^^^ Sr -"---^ his 
 
 *afe 6ee» descried reached flTZ f'^^'^O }1ie precipices which 
 
 whole march, many^f tf folLf frl hJstl SeL"™"^ f 
 crossmg rivers, and many also in thT Alt? ^' '*"'' ''^ 
 
 precipices and other difficultts of the rou'tt^^JeTeS"' *'^. 
 horses and ba2<>-a<'e-animnl« fi.^^ tk T ™"'^®' wmle the loss of 
 
 than that of "men Tit It rt^' *'* """'"' ^'^^ «"" S^^^^ter 
 
 ^.r^^ OF THE Po, AND THE COUNTRY OP THE IvsUBRTAivr^i nP U- 
 
 Libyan forces, 12,000 infantry were nil fhof ^^^^Y^'^J^®- ^f his 
 
 Chap. GO. 
 
 o< <Ae wr« foot of the Alnl h» o T ^' '"^''''^ encampet/ 
 
 the stren^^th of {if tro^s. FoTleTht/'"' '" T* "■•^= 
 only suffered very severely durin°f.* ^"^J" T"^ ^^^ "»* 
 in the rugged country abo^t the f„mmit o/ tL"'' '"f"'' "*"'• 
 were also in a wretched conditionTom\h1 waTo^f ^rolS 
 
 ( 
 
 / 
 
 / i 
 
 I 
 
 -r 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 15 
 
 ♦they had experienced, and the unavoidable neglect of their 
 bodily health. Many indeed were depressed to the lowest state 
 by the privations and incessant hardships they had undergone. 
 For it was impossible to convey food in quantities sufficient for 
 so many thousand men, through a country which presented such 
 difficulties : and even of the provisions which they did convey 
 with them, the greater part was lost when the beasts which 
 carried them perished. And thus, having begun his march, after 
 the passage of the Rhone, with 38,000 infantry, and more than 
 8000 cavalry, he lost, as I have before mentioned, nearly half his 
 whole force in the passage of the Alps ; and even those who did 
 survive, had become from incessant hardship quite savage in ap- 
 pearance and condition. Hannibal, therefore, used his utmost 
 efforts to revive the spirits of his army, and to restore the strength 
 of both men and horses. When the army had now completely 
 recovered, Hannibal tried to effect an alliance with the Taurini, a 
 nation dwelling at the foot of the mountains, who were at Avar 
 with the Insubrians, and distrusted the Carthaginians. Upon their 
 rejection of the offers of Hannibal, he invested their principal 
 city ; and, having taken it after three days' siege, he put to the 
 sword all that had opposed him, and inspired such terror in the 
 minds of the neighbouring barbarians, that they all came imme- 
 diately and threw themselves upon his protection. The remain- 
 ing tribes also of the Celtae, who dwelt in the plains, were very 
 anxious to co-operate with the Carthaginians according to their 
 original intention ; but the Roman legions having already passed 
 through the country where the greater part of them dwelt, and so 
 cut them Oif (from the Carthaginians), they either remained neu- 
 tral, or were even obliged in some cases to join the Roman army. 
 When Hannibal considered these things, he determined to lose 
 no time, but to advance forward and achieve some success, which 
 might serve to encourage all those who were willing to share the 
 fortunes of the Carthaginians. 
 
 1-, 
 
 i 
 
■PVmBPnp 
 
 
 lii 
 
 1 ' 
 
 l\ 
 
 CHAPTER III. 
 
 Discussion of the narrative of Polybius, from the passage of the 
 Rhone to the commencement of the ascent of the Alps. — Iden- 
 tification of the Scaras of Polybius with the modern Isere. 
 
 The Island of Polybius the district between the Rhone and the 
 Isere.— Polybius only partially acquainted with this district. — 
 Conclusion drawn that Hannibal must have marched up the left 
 bank of the Isere, and not up the Rhone, from the junction of 
 the two rivers. 
 
 mHE course of Hannibal's march from New Carthao-e 
 J- has been already traced. After crossing the Pyrenees 
 and the Ehone, he proceeded upwards along the eastern 
 bank of that river until he came to a district called the 
 Island, which Polybius compares in form and size to the 
 Egyptian Delta. One side of this triangular district was 
 formed by the Rhone, another by the confluent stream of 
 a river called the Scaras, and the third by a range of almost 
 inaccessible mountains extending across from the Scaras 
 to the Rhone. The first division of the march from the 
 passage of the Rhone to the plains of Italy terminates at 
 this district, and must of course be supposed to end at the 
 point where Hannibal first came upon its borders : for that 
 would be the place where he would naturally be said to 
 arrive at it, besides being the only definite point where 
 the march could be broken off: Now. as Hannibal 
 inarched along the bank of the Rhone, which formed one 
 side of the Island ; and as the Scaras, a river falling into 
 the Rhone, formed its second side ; it must have been at 
 the junction of these two rivers that the account of Han- 
 nibal s march is broken off by Polybius ; and it must be 
 from the same place that he afterwards resumes it, when 
 he has completed the narration of the events which took 
 place at the Island. The first point therefore to determine 
 IS, what was this river which Polybius calls the Scaras 
 
 , / 
 
 I 
 
 \ 
 
 ,1 
 
 ( 
 
 If 
 I 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 17 
 
 Now the name ,Scara,s is unknown in Roman geo- 
 graphy, but may be shewn to be virtually identical with 
 Isaras; and the river Scaras must consequently be the 
 Roman Isara, the modern ^ere. The word Isara or Isaras 
 iS plainly a compound word, the elements of which are 
 Is-aras. Of the derivation of its latter element aras there 
 can be little doubt. It is of the same origin as the Roman 
 Arar, (the Saone) which is a reduplication of the syllable ar, 
 one of those syllables consisting either of the vowel a alone,' 
 or of the same vowel combined with various consonants,' 
 which form, either singly or reduplicated, the appellations 
 of water and rivers in many languages, both of the East 
 and the West. There might be cited, as examples, the 
 Persian a6, the Latin aqua, the Ach or Ache of the Aus- 
 trian Alps, the Swiss or French Aa, the Swiss Aar, the 
 Anas of Boetica, and many other words of similar formation. 
 The Is of the word Isaras has however another origin, and 
 is identical with the Celtic uisg or uisge " water," a term 
 which is still applied to some of the streams of the Scotch 
 Highlands, and from which, by corruption, such names of 
 rivers as Ouse, Usk, Esk, Exe, &c., are derived. Of the 
 Celtic uisg, the Romans, as in the case of the Usk, formed 
 the word Isca, which may thus be taken as the classic 
 form of the Celtic word. The two elements therefore, which 
 enter into the name Is-aras are Isca and Aras; which, 
 when blended into one word, would give Iscaras as' the pri' 
 mitive form of Isaras. By the omission of the initial vowel, 
 the Greek word Scaras at once results ; while the Latin 
 Isara is clearly but a more euphonious form of Iscaras, We 
 may thus conclude that Scaras and Isara are both merely 
 different forms of one word, Iscaras, and that the river 
 Scaras of Polybius is the modern Isere. The Isca and the 
 Aras would, in all probability, be the names of the two 
 principal rivers which united to form the Iscaras ; the Isca 
 being upon this supposition identified with the Upper Isere, 
 the river of the Tarantaise ; and the Aras with the Arc,' 
 the river of the Maurienne\ 
 
 The distances given by Polybius are in perfect accord- 
 
 1 Similar derivations of the names of rivers are given by Dr Donald- 
 BOn in his VarronianuSy p. 45, (2nd Edit.) 
 
«■ 
 
 l^eHSrsmmmm 
 
 iqpp* ^ 
 
 18 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 ance with the conckision that the Scaras and the Isere are 
 identical. From the point where Hannibal passed the 
 Ehone to the "ascent of the Alps," the length of the march 
 was 1400 stadia. From the Island also, that is to say, 
 from the junction of the Rhone and the Scaras, there war 
 a march of 800 stadia to the same point. There was there- 
 fore a distance of 600 stadia or 75 Eoman miles from the 
 point where the Rhone was crossed, to the junction of the 
 Scaras and the Rhone. The place where the Carthaginians 
 passed the Rhone was also at a distance of about 1600 
 stadia or 200 Roman miles from Emporium, and nearly 
 four days' march from the sea. 
 
 These distances are in themselves almost, if not quite, 
 sufficient to prove, that Hannibal crossed the Rhone at or 
 near Roquemaure, and that the Scaras of Polybius was the 
 Istre. The question has been argued satisfactorily by M. 
 Deluc upon these grounds ^ Following the Roman itine- 
 raries as far as possible, he gives the distance between the 
 ancient Emporium and Roquemaure at 206.2 Roman miles; 
 the distance between the eastern embouchure of the Rhone 
 and a point on that river opposite Roquemaure at 64j 
 Roman miles ; and the distance from this latter point to 
 the junction of the Rhone and Isere at 75 Roman miles. 
 The last of these distances is accurately the same as the 
 distance of 600 stadia, deduced from Polybius as the 
 length of the march from the passage of the Rhone to the 
 confluence of the Scaras. The actual 206.2 Roman miles 
 between Emporium and Roquemaure make also a distance 
 nearly in accordance with that given by Polybius, who 
 states it as being about 1600 stadia or 200 Roman miles. 
 Finally, the distance of 64 1 Roman miles, from the point 
 on the Rhone opposite Roquemaure to the eastern embou- 
 chure of the Rhone, would be a distance that might be 
 well described as rather under four days* march, as Han- 
 
 1 Histoire du passage des Alpes par Annibal. Ukert conjectures 
 that Hannibal crossed the river lower down, in the neighbourhood of 
 Beaucaire ; a supposition which is also adopted in a very able article 
 in the Philological Museum, on Hannibal's passage of the Alps. On 
 many material points, however, as will be afterwards found, the views 
 advocated in this treatise receive the sanction of the learned writer of 
 the article alluded to. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 19 
 
 I 
 
 nibal, after having passed ihe Rhone, marched to the con- 
 fluence of the Scaras, a distance of 75 Roman miles, in 
 four complete days. 
 
 The description which Polybius gives of the district 
 called the Island, though not minutely accurate, is j^et 
 sufficiently so for its recognition. Two sides of it, he says, 
 were bounded by the Rhone and the Scaras : its third side 
 was formed by a chain of almost impracticable mountains. 
 These mountains spoken of by Polybius form what is 
 called the Chartreuse range, the celebrated monastery of 
 the Grande Chartreuse being inclosed among them. They 
 extend northward from Grenoble to the outlet of the lake 
 of Bourget, and form the connecting link between the 
 Alps and the Jura, though not strictly belonging to either 
 mountain-system. On the east, they are separated from 
 the Alps by the Vale of Graisivaudan, (the French part of 
 the valley of the Isere) by the valley of Chambery, and by 
 the lake of Bourget. On the south-west, they are also 
 separated from the Alps by a portion of the Vale of Graisi- 
 vaudan. On the north-west, the Rhone divides them from 
 the Jura : and on the west, the river Guiers, which marks 
 the frontier between France and Savoy, and a line drawn 
 from Les Echelles to Moirans, may be assigned as the 
 limits between them and the hilly part of the district 
 between the Rhone and the Isere. The character which 
 Polybius gives these mountains is perfectly accurate. 
 Their extreme western face rises suddenly and precipi- 
 tously from the country they overlook, and presents a long 
 barrier of rock so apparently impracticable, as fully to bear 
 out Polybius' strong expressions, that these mountains 
 were difficult of approach and entrance, and almost inac- 
 cessible. 
 
 Yet, though the Island of Polybius may thus be iden- 
 tified by the nature of the boundaries he has assigned to it, 
 his acquaintance with it does not seem to have been per- 
 fect. The comparison of it to the Egyptian Delta is not 
 very fortunate. It is considerably less in size than the 
 Egyptian district, and does not entirely resemble a Delta 
 in form. For, in order that this comparison might hold, 
 the course of the Rhone, from the confluence of the 
 
 2—2 
 
 >> / 
 
20 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS, 
 
 21 
 
 Isere to the foot of the Chartreuse mountains near 
 St Genix, ought to be straight. Yet the course of this 
 river (reckoning upwards) is only straight as far as Lyons. 
 It there changes its direction from north to east, assumes 
 subsequently, for a short distance, a north-easterly course, 
 and finally makes a sudden flexure towards the south-east, 
 retaining the same direction as far as St Genix. The 
 only way of accounting for this defective description of 
 the Island seems to be, by supposing Polybius not to have 
 been personally acquainted with the northern parts of that 
 district. It is true that he would have no maps to assist 
 him in forming a conception of the country, but could only 
 judge of its figure and nature by the eye. Yet, even with 
 these disadvantages, it hardly appears probable that he 
 could have been ignorant of the flexures of the Rhone 
 above Lyons, or could have looked upon it as running a 
 straight course. The length also of the base of the Char- 
 treuse mountains, extending from Moirans to St Genix, is 
 much too short to allow them to be cited as standing in 
 the same relation to the Island that the sea does to the 
 Egyptian Delta. If, however, we suppose Polybius only to 
 have visited the southern parts of the Island, the resem- 
 blance which he finds between it and the Egyptian Delta 
 is readily understood. The Isere does actually run nearly a 
 straight course from the edge of the base of the Chartreuse 
 mountains near Moirans down to its confluence with the 
 Ilhone, and therefore corresponds perfectly to the eastern 
 arm of the Nile. The course of the Hhone also from 
 Lyons downward may be considered straight, and will thus 
 represent the western branch of the Nile, if its course 
 above Lyons be supposed unknown, and the river be 
 imagined, as was probably the case with Polybius, to pre- 
 serve the same direction above that city which it does 
 below. Upon this supposition, the Ilhone above Lyons 
 would be conceived by Polybius, not as penetrating from 
 the lake of Geneva between the Jura and the Chartreuse 
 mountains, but as lying on the west, instead of the east, 
 of the Jura range. The mountains therefore which Poly- 
 bius speaks of as forming the third side of the Island, 
 would not now be merely the Chartreuse range, but woidd 
 
 (I \ 
 I 
 
 
 include, besides, the extremity of the Jura, extending as 
 a continuation of the Chartreuse mountains from the 
 Ehone to the Ain, and thus nearly reaching the ima- 
 ginary prolongation of the Rhone above Lyons. Both 
 these systems of mountains would appear from the country 
 of the Island as a connected group, and would naturally 
 be considered to form as such one of the boundaries of the 
 district. 
 
 Upon the supposition therefore that Polybius had not 
 visited the northern parts of the country between the 
 Rhone and the Isere, the discrepancies in the imagined 
 resemblance between the Island and the Egyptian Delta 
 may be accounted for ; for the similarity holds good to the 
 eye, if the course of the Rhone above Lyons be supposed 
 unknown, and the Jura and Chartreuse mountains be taken 
 together as one mountain boundary. That Polybius also 
 exaggerates the size of the Island is another argument 
 that he was not fully acquainted with it. Yet, if he was 
 ignorant of any district in it, that district must have been 
 the northern or more remote portion of the Island : for, 
 if he was acquainted with those parts of it, he could not 
 have been ignorant of the southern parts. That the Island 
 was the country between the Rhone and the Isere, ap- 
 pears clearly, as has been previously shewn, from the dis- 
 tances given by Polybius, and from the identity of the names 
 Scaras and Isaras. The defective resemblance between 
 this tract of country and the Egyptian Delta does not 
 invalidate this fact, but only leads us to conclude that 
 Polybius had not visited the northern parts of the country; 
 a conclusion which will be found of some importance in 
 the investigation of the route by which Hannibal reached 
 the first ascent of the Alps ^ 
 
 1 It may be here observed, that this country, though not conforming 
 accurately to the shape of a Delta, but only taking that form in the mind 
 of one not thoroughly acquainted with it, is yet exceedingly well described 
 by the native name of *the Island/ With the exception of the line 
 drawn from the Isere near Moirans to the Rhone near St Genix, a line 
 hardly exceeding 20 miles in length, those two rivers form its entire 
 boundary Consequently, out of a circumference of nearly 180 miles, it 
 is surrounded for about 160 miles, or eight-ninths of the whole circuit. 
 
 »■ # 
 
-r^ 
 
 22 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 I' 
 
 #1 
 
 ! ! 
 
 I 
 
 .1 
 
 This route is the subject which has now immediately 
 to be considered. After having established the elder of 
 the two contending brothers in the sovereignty of the 
 Island, Hannibal resumed his march under the escort of a 
 band of the men of that district*. This march must be 
 supposed to be resumed from the point where Polybius 
 breaks off his narrative of the previous march ; from the 
 confluence, that is to say, of the Ehone and the Isere. 
 And here a difficulty at once arises to be resolved : for the 
 resumed march is said to have lain bj/ the river side for a 
 distance of 800 stadia or 100 Roman miles. AVhich then 
 of the two rivers, the Rhone and the Isere, from whose 
 junction he started, was the river which Hannibal followed 
 up? 
 
 The previous narrative of Polybius would at first lead 
 us to imagine that it was the Rhone. For in the 39th 
 chapter it is stated, that, after the passage of the Rhone, 
 Hannibal went along the very river bank towards its source,' 
 until he arrived at the ascent of the Alps on the way to 
 Italy. Now from this account it is certainly most natural 
 to conclude, that it was the same river which was followed 
 up all the way to the beginning of the Alpine ascent ; and 
 not in the first part of the march the Rhone, and in the 
 second the Isere. Another apparent argument in favour 
 of the Rhone is, that Hannibal's operations in the Island 
 must have obliged him to cross the Isere, and led him 
 away from the banks of that river. This last argument is 
 however of no weight. The transactions in the Island are 
 quite episodical to the rest of the march, and seem to be 
 kept perfectly distinct by Polybius, who breaks off his nar- 
 rative of the progress of the army, for the purpose of 
 relating them, at the junction of the Rhone and the Isere, 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 23 
 
 by tho waters of two large and navigable rivers. Under these circum. 
 stances, and considering that it is an inland district, the name of * Island/ 
 as applied to it, is a very expressive and appropriate designation. 
 
 1 The Island afterwards formed part of the country of the Allobro- 
 ges. This does not, however, appear to have been the case in the time 
 of Hannibal, for Polybius speaks of * the Gauls called Allobroges,' as if 
 they were entirely a different people. See Philological Museum, * Han- 
 nibal's Passage of the Alps/ 
 
 \ 
 
 as if to intimate that the bulk of the army went no further 
 northward at that time along the Rhone. The transactions 
 at the Island stand indeed, in their relation to this part of 
 the march, quite in the same position as the subjection of 
 the Spanish nations between the Ebro and the Pyrenees 
 does to the march along the coast in that part of Spain. To 
 subdue these nations, Hannibal must have left the coast, 
 and penetrated into the interior of the country ; and yet, 
 in spite of these inevitable deviations, Polybius makes 
 Hannibal's march to lie throughout along the coast of 
 Spain from New Carthage to Emporium. Indeed, had he 
 adopted any other system, great confusion must have been 
 the result, and all estimation of distances nearly imprac- 
 ticable. All the marchings and countermarchings there- 
 fore, which Hannibal performed in the course of his opera- 
 tions against these Spanish tribes, from the time he left 
 the coast-road until the time that he returned to it, are 
 entirely omitted by Polybius in his relation of the progress 
 of the great march. The transactions at the Island seem 
 to be treated by him in the same manner. Before pro- 
 ceeding to relate them, he arrests the march of the army 
 at the confluence of the Isere with the Rhone, measuring 
 the distance up to that point. No distances whatsoever 
 are mentioned while the operations in the Island are 
 related. After concluding his account of those events, the 
 estimation of distances again begins ; and this estimation 
 of distances must clearly be commenced, as has already 
 been often repeated, at the place where the last measure- 
 ment of distances terminated, or at the junction of the 
 Rhone and the Isere. 
 
 The first argument against the supposition that Han- 
 nibal followed up the Rhone from the junction of the two 
 rivers is derived from the fact, that the whole Carthaginian 
 army must in that case have passed the Isere, and that the 
 passage of that river is not mentioned by Polybius. It is 
 true that, in order to carry on his operations in the Island, 
 Hannibal must have crossed the Isere with part of his 
 army ; but there is no sufficient reason for concluding that 
 the baggage and the elephants were then brought across. 
 Had they been so, the silence of Polybius on the subject 
 
ms 
 
 24 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 '• \ 
 
 ) 
 
 ! I 
 
 ', ! 
 
 \ 
 
 > 
 
 would be difficult to account for. There were indeed 
 several rivers crossed by Hannibal in Gaul, of which Poly- 
 bius takes no notice ; but these rivers would all be small 
 streams, which the army would pass by wadinff. The 
 Isere cannot be ranked in this class. It is. at its junction 
 with the Rhone, a navigable river with a deep and rapid 
 current, and presents an obstacle to the passage of an 
 army too important to be omitted in silence in a narrative 
 such as that of Polybius. With the exception of hostile 
 opposition on the part of the inhabitants, all the difficulties 
 and dangers which attended the passage of the incum- 
 brances of the army across the Rhone, must have been 
 repeated m the case of the Isere. the lesser width of the 
 latter river being the only circumstance that would make 
 any difference. And yet no word about difficulty or dan- 
 ger is spoken by Polybius with respect to the Isere, nor 
 
 to'TfFlV.l "''*'*'*' ''^/"/ preparations having been made 
 to eltect the passage of the river. 
 
 Suppose, however, Hannibal to have brought all his 
 army across the Isere, and to have directed his mareh 
 upon the Alps along the side of the Rhone. In this ease, 
 the firs mountains that he would encounter upon his wa^ 
 would be the Chartreuse group, the foot of which he 
 would reach when he arrived in the neighbourhood of 
 bt Genix, at the junction of the Guiers with the Rhone. 
 St Genix would thus be the earliest place, where the first 
 ascent of the Alps could be fixed. 
 
 But the first ascent of the Alps lay at a distance of 
 100 Roman miles from the junction of the Rhone and the 
 Isere reckoned along the river side. St Genix should 
 therefore he at about that distance from the confluence of 
 
 LTof'lT"' ^^T ''"" ^^"^^"^^ ^^ ^^^«« the Itine- 
 rary of Antonmus reckons 71 Roman miles. Allowing 6 
 
 Roman miles for the distance from the confluence of the 
 
 Rhone and the Isere to Valence, there remain 65 for the 
 
 distance to Lyons. Above Lyons there are no ancient or 
 
 modern roads along the Rhone to assist us in determining 
 
 length between Lyons and the confluence of the Isere, and 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 25 
 
 we may therefore take the probable length of a march 
 along the rivor as being the same in both cases. The 
 distance from the junction of the Rhone and Isere to 
 St Genix, reckoned along the Rhone, would thus be about 
 130 Roman miles, instead of the 100 given by Polybius, a 
 discrepancy which seems greater than ought to occur, 
 though perhaps not so great, considering the manner in 
 which Polybius measures, as to be decidedly conclusive 
 against the supposition of HannibaPs having followed up 
 the Rhone*. 
 
 It has already been conjectured, in discussing Polybius' 
 account of the Island, and in endeavouring to account for 
 his imperfect conception of the nature of that district, that 
 he had not visited its northern parts. If such were the 
 case, it would be a strong argument that Hannibal never 
 followed up the Rhone to the foot of the mountains ; for 
 otherwise Polybius, who had examined the whole of Han- 
 nibaPs route, must have visited the northern parts of the 
 Island, and have thus been enabled to form a more accu- 
 rate idea of its nature and figure. 
 
 It may also be mentioned, that the very cursory man- 
 ner in which Polybius speaks of the Chartreuse mountains 
 in his description of the Island, and his not applying to 
 them the name of Alps, would hardly lead us to expect 
 that he considered them as forming the very portion of 
 that great mountain system which Hannibal first encoun- 
 tered. We should rather imagine that he looked upon 
 them as mountains independent of the Alps, and with 
 which he had no further concern, than in as far as they 
 formed one of the boundaries of the district called the 
 Island. 
 
 1 M. Deluc, who supposes that Hannibal brought his whole army 
 into the Island, and subsequently crossed the Chartreuse mountains, 
 makes him leave the Rhone at Vienne, and strike straight across the 
 country to St Genix. It seems however impossible, that, if Hannibal 
 had adopted this route, Polybius could have spoken of him as keeping 
 along the very bank of the river to the foot of the mountains ; or that 
 the sharers in the expedition, from whom the accounts of it must have 
 been derived, should have failed to observe, that they had turned their 
 backs upon the Rhone at Vienne. 
 
 /i 
 
26 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 Another objection to the supposition of HannibaPs 
 having followed up the Rhone arises from the fact of the 
 extremely circuitous nature of the route, from the junction 
 of the Rhone and the Isere to St Genix, which the bank 
 of the Rhone affords. It seems indeed nearly twice as 
 long as is requisite. This objection might possibly appear, 
 at first sight, somewhat obviated by the consideration, 
 that the country was not then accurately mapped out, as it 
 is now, and therefore the distances between various points, 
 and their respective bearings, w^ere not sufficiently well 
 known. But it must be remembered, that Hannibal would 
 then be under the escort of the inhabitants of the country 
 he was traversing, whose local knowledge would compen- 
 sate their want of geographical information: and it is 
 almost impossible to conceive that they, wishing to guide 
 the Carthaginians from the junction of the Rhone and 
 Isere to St Genix, should have conducted him there all 
 the way along the very bank of the Rhone. Yet this, 
 according to Polybius' account, they must have done, if 
 the Rhone were indeed the river which Hannibal followed 
 to the foot of the Alps. 
 
 Again, if Hannibal followed up the Rhone to St Genix, 
 he would either subsequently continue to pursue its banks 
 to the lake of Geneva, and enter Italy by the Simplon or 
 the Great St Bernard, or else cross the Chartreuse moun- 
 tains, and make for the Little St Bernard or the Mont 
 Cenis. Now that Hannibal entered Italy by the Simplon 
 or the Great St Bernard are suppositions so beset with 
 improbabilities that it is hardly necessary to consider them : 
 so that the only point which it is requisite to examine here 
 will be, whether any additional improbability arises from 
 supposing that Hannibal selected the banks of the Rhone 
 as his line of approach to the pass of the Little St Bernard 
 or of the Mont Cenis. 
 
 Now, if he chose this line of approach, he must cross 
 the Chartreuse mountains, and subsequently pass through 
 Montmelian on the Isere. Yet, if a general wished to 
 reach Montmelian from the junction of the Rhone and 
 Isere, it would be very much shorter for him to follow up 
 the Isere, rather than to take the line of the Rhone, and 
 
 i i 
 
 t 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 27 
 
 cross the Chartreuse mountains. Montmelian is, m fact, 
 by the high road along the north bank of the Isere, only 
 about 100 Roman miles from the confluence of the Rhone 
 and Isere, while the distance along the banks of the 
 Rhone and across the Chartreuse mountams exceeds 160. 
 If then the longer line was adopted, it would naturaUy be 
 chosen because it was a much easier route. But the Ime 
 of the Isere is not only the shorter, but at the same time 
 the easier route ; for in the other case the Chartreuse 
 mountains have to be crossed, a range of mountams which 
 Polybius mentions as almost impracticable. It is very dif- 
 ficult to imagine that Hannibal would lengthen this part of 
 his journey by more than one half, not in order that he 
 might avoid, but in order that he might cross without 
 necessity, mountains described by Polybius himself, as 
 
 /I 
 
 From all these considerations the natural conclusion 
 seems to be, that, although the most obvious interpreta- 
 tion of the words ' ^apd t^v TroTaf^ov,' m chap. 39, would 
 lead us to think that Hannibal followed up the Rhone, from 
 the place where he crossed that river up to the first ascent 
 of the Alps ; yet the Isere must nevertheless be taken as the 
 river which led Hannibal to the Alps after the transactions 
 in the Island, and we must suppose that he followed up the 
 Rhone no further than to the point where the Isere joins 
 it The passage of the Isere not being mentioned by Poly- 
 bius in his account of the march of the army to the Alps, 
 v.e shall infer that one part of the army with the e le- 
 phants and baggage, remained on the left bank of the 
 Isfere while Hannibal crossed the river with the rest to 
 take 'part in the affairs of the Island. After the conclusion 
 of these transactions, Hannibal would recross the Isere, 
 and the whole army would be again collected together on 
 the south of the river, and at the place were it flows into 
 the Rhone. From this point the march to tlie Alps wan 
 resumed. It lay, as Polybius informs us, m the begmning 
 of his 50th chapter, for 800 stadia along the river side. 
 
 1 See TJkorfs Geographk, and Philol. Mm. Vol. il. 
 
 -ssi 
 
 
L 
 
 
 ill 
 
 I 
 
 • V.i>iWP i P»M>-,i*^" 
 
 \) 
 
 28 
 
 HANNIBAL*S PASSAGE OF THE ALPS. 
 
 The ascent of the Alps (or to the Alps, as Polybius here 
 expresses it) then commenced ; that is to say, the route, 
 previously upon the plain, (eV Toh eTnireU,) now first 
 began to ascend the mountains. Here then, having previ- 
 ously concluded that iha Carthaginian army never "^crossed 
 the Isere on their way to the Alps, we have another con- 
 dition for the determination of Hannibal's route : 
 
 * The commencement of the ascent of the Alps must be 
 situated at a distance of about 100 Eoman miles from the 
 junction of the Khone and the Isere, reckoned along the 
 left bank of the latter river ^' 
 
 -II, 
 
 Condition I. 
 
 y 
 
 l' CHAPTER IV. 
 
 Discussion of the narrative of Polybius, from the commencement of the 
 ascent of the Alps to the arrival at the summit of the pass. — The 
 defile at the commencement of the ascent. — Conclusions drawn as 
 to its nature and the character of the neighbouring country. — The 
 'strong white rock.' — Features of the country in its vicinity. — The 
 * rock' at a distance of many miles from the summit of the pass. 
 
 WHEN the Carthaginians had approached the com- 
 mencement of the mountain route, the men of the 
 Island returned to their own country, and left them to 
 penetrate alone into the Alps. They encountered almost 
 immediately the greatest dangers. A difficult pass, 
 through which they were obliged to march, was com- 
 manded by the Allobroges, who had taken possession of 
 the heights above the road. This pass was situated, it is 
 clear, at or very near the commencement of the ascent. 
 Its nature, and the features of the surrounding country, 
 will now have to be investigated. 
 
 The first features presented to our notice are certain 
 ^ advantageous posts,' which Polybius afterwards mentions 
 as * heights.' The elevation of these heights was evidently 
 considerable, and their access difficult, at least from the 
 side where Hannibal lay with his army : for the Allobroges 
 would hardly have ventured to abandon them during the 
 night, had they not then thought them perfectly secure 
 from occupation on the part of the Carthaginians. It is 
 also mentioned, that when Hannibal set forward to seize 
 them in the night, the soldiers whom he selected for this 
 service were active and lightly equipped, in order that 
 they might be better enabled to contend with the difficul- 
 ties of the mountains. It appears, besides, from the 
 account of the battle of the next day, that Hannibal's posts 
 on the heights were at some distance from the road 
 through the pass, along which his army was defiling when 
 
 i^... 
 
 w 
 
 kl 
 
so 
 
 HANNIBAL'S PASSAGE OF TlIK ALPS. 
 
 UANNICAL'S PASSAGE OF THE ALPS. 
 
 31 
 
 M' 
 
 Jt WHS iitlackea by the jMlobrofrcs. ].•„,. „s ^U^, Ca.il.an.i. 
 luans were skirting the edge of the precipices uhicli lay 
 below the road, and Hannibal was in position on the i 
 heights, the Allobroges attacked the army on its march j 
 from the ground above the road, and placed themselves 
 between Hannibal and the rest of the Carthaginians. This 
 IS clear from wliat subsequently followed, when Hannibal 
 at length charged down from the heights and routed the I 
 Allobroges,) destroying an equal number both of his owa) 
 army and of the enemy : for the Carthaginians, bein-^ ^ 
 below the Allobroges, were forced over the precipices by 
 the success of the attack upon the enemy above them, who 
 were driven down in confusion towards the preci.iices 
 Isow, had there not been a considerable interval between i 
 Hannibal and the rest of his army, the Allobroges would 
 scarcely have dared to occupy the intervening space, and 
 expose themselves immediately to attack from Hannibal i 
 and this before they had any time to inflict much loss \ 
 upon the Carthaginian army. AVith respect then to these 
 heights the inference seems to l)e, that they were of con- 
 siderable elevation, and not readily accessible from tlie side 
 where the Carthaginians lay, though probably more easily 
 gamed from the side of the Allobroges, a circumstance 
 which would induce them to abandon their posts durln- 
 the night with less apprehension. 
 
 When Hannibal seized in the night the posts which had 
 been abandoned by the Allobroges, he first passed through 
 r« <rre.a ' the defile.' There was then, as might have been 
 otherwise concluded, a dclile in this part of the route, and 
 Hannibal was obliged to pass through it on his inarch to 
 seize the heights. 
 
 Polybius describes the May by which the Carthaginians 
 had to advance, as being narrow and rugged, and also 
 precipitous. A great number of horses and ba-^ra^c! 
 animals are said besides to have been lost <lown theTre^ci 
 pices, when the Allobroges attacked the Carlha-ii.ians 
 lu some part, therefore, of the line of advance, a ".arrow 
 and rugged w-ay along the edge of a precipice must be 
 found for the Carthaginians. 
 
 -Two other points which demand attention arc sn-. 
 
 !*•'' 
 
 ♦ I- 
 
 gcstcd by ilui liiftlory of ilio events wliicli inc'ccdcMl mid 
 followed the battle at tlic delile. On tlie evening of the 
 day before the battle, Hannibal advanced into the imme- 
 diate neighbourhood of the defile, and encamped not far 
 from the enemy, their position being on tlie heights or 
 advantageous posts. Previously also to his setting out in 
 the night to seize these i)osts, he conunandcd tlic camp 
 fires to be lighted ; no doubt >vitli the intention that the 
 Allobroges might sec them, and, concluding that the 
 Carthaginians were permanently encamped till the morn- 
 ing, and meditated no further movement for the present, 
 might, according to their custom, leave their posts for the 
 night, and retire to the neighbouring town. The truth 
 of this supposition is confirmed by its being said, that 
 Hannibal advanced to the i)lace where he encamped, in 
 open view of the enemy. Hicre must consequently be 
 found, immediately below tlie defile, a place of encamp- 
 ment for Hannibal's army visible from the heights com- 
 manding the defile. 
 
 The town to which the Allobroges retired for the 
 night presents a second subject for consideration. This 
 town would be situated above the defile, and not far from 
 it : for it is not likely that the Allobroges would have 
 retired to any considerable distance from their strong 
 posts. Indeed, Polybius expressly describes it as a town 
 in the neighbourhood of the heights. Above the defile 
 therefore, and near it, must be found a place where a 
 town either stands or might have stood. ]kit more than 
 this is even requisite: for Hannibal, having taken the 
 town, encamped there for a whole day. The place sought 
 must thus be sufilciently large, not merely for the site of a 
 town, but also for the encami)mcnt of an army as nume- 
 rous as Hannibal's, which at this time probably amounted 
 to more than 40,000 men, besides elci)hants, cavalry- 
 horses, and beasts of burden. An open district or valley 
 among the mountains seems to be required in this iilace. 
 
 Such then are the characteristics of the defile and the 
 adjacent country, which the narrative of Polybius suggests 
 to us. The existence of such a defile supjilics another 
 important clue by which Hannibal m:iy be traced in his 
 
 \i 
 
 
 
 ''■if?1 
 
 ■' ''"r ij"M.Ai..a,i>li..m^?r:-~igiBs;i.-- 
 

 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 33 
 
 4 < 
 
 
 I ] 
 
 4 t. 
 
 K t "' 
 
 .1 '} 
 
 Alpine route; the cliaractcr of the pass and the nature of 
 the vicnnty presenting several particular features by which 
 It may be identified. AVe thus derive, from the seetion 
 of Polybuis under consideration, one principal condition, 
 as well as several minor ones, which any route, supposed 
 to be that of Hannibal, must satisfy. The principal con- 
 V dition may be stated as follows » : 
 
 ^ ' At the commencement of the ascent of the Alps a 
 defile nuist be found, the character of which, and of the 
 surrounding country, is in accordance with the events 
 related by Tolybius to have occurred in that place.' 
 
 The minor conditions which this includes may be thus 
 given : 
 
 (1) The defile must be commanded by certain heights, 
 ot considerable elevation. 
 
 (2) These heights must not be easily accessible from 
 below the defile. 
 
 (3) The way through the defile must skirt in some 
 place the cdrrc of a precii)ice. 
 
 (4) Immediately belo;v the defile, must be found a 
 place where an army as numerous as that of Hannibal 
 could encamp. 
 
 (5) This place of encampment should be visible from 
 the heights commanding the defile. 
 
 (G) Above the defile, and near it, must be found an 
 open district, where a town either stands or mi^rht have 
 stood, and wlicrc an army such as Hannibal's, could have 
 \ encamped. 
 
 The ten days^ march from the junction of the Rhone 
 and Isore must be taken as terminating, not at the point 
 where Hannibal left the Iscre, but at th^ town of the Alio! 
 broges, which he captured after passing through the defile 
 at the commencement of the Alpine ascent. For from 
 this town the subsequent march of m^^^n days through 
 
 otht fo T ""'^ "^'""' •• ^"^^ '''^ ^--^^ -'^ ten days 
 oiiglit to terminate where the march of fifteen days be- 
 
 J Condition HI. 
 
 gins. The distance also of 100 Homan miles along the 
 river side is hardly of sullicicnt length to have occupied 
 the Carthaginians ten days in marching ; for even in the 
 Alps their rate of advance was greater than this. It 
 would be on the evening of the eighth day after leaving 
 the confluence of the Ehone and Iscre, that Hannibal en- 
 camped before the heights occupied by the Allobroges : 
 on the same night he seized the abandoned heights: on 
 the ninth day the defile was j)assed, and the town cap- 
 tured : and on the tenth day the Carthaginians remained 
 encamped in the neighbourhood of the town. Polybius, 
 indeed, estimates the length, in distance, of the passage 
 of the Alps, from the point where Hannibal left the Isere ; 
 so that it might be natural to expect that the fifteen days 
 occupied in that ])assage would be reckoned from the 
 same point. I3ut, from the rest of the narrative, it seems 
 plain, that the fifteen days' march is reckoned from the 
 town, the caj)ture of which makes a natural break in the 
 history. This town was also, no doubt, very near i\\Q 
 Iscre, and thus only a short distance removed from the 
 point where the march along the river terminated. 
 
 The fifteen days'* march through the Alps terminated ^^ 
 at the commencement of the great plains of Northern 
 Italy. (Chap. 39). On the ninth day after leaving the 
 town, the summit of the pass was gained ; and the army 
 remained encamped there for two days. The descent 
 then began ; but was soon arrested in consequence of 
 the destruction of the road, by which the army had to 
 descend. It was only after three days of hardship that 
 this obstacle was surmounted, and the way made practi- 
 cable. In one day the ])ath was sufiiciently repaired for 
 the passage of the horses and baggage-animals; but it 
 was not till the third dny that the elephants were able 
 to pass. On the third day after leaving the place where 
 the path was broken uj), the commencement of the plains 
 of Italy was reached. ^y^ 
 
 Since the whole march through the Alj)S occupieclv 
 fifteen days, and the march terminated on the third day 
 after IIanni])al ])assed the ])lace where the ])ath was bro- 
 ken up, it must have been on the thirteenth day that thia 
 
 3 
 
 1 
 
 
 : 1 
 
 i 
 
 Li ■' ' 
 
 
 
-:i 
 
 I 1 
 
 S-i IIANxXIBAL'S PASSAGE OF THE ALPS. 
 
 latter event took i)lacc. Ikit tlic Cnrthn^IniMns were dc- 
 taiuctl for three days by tlie broken path, and conse- 
 qucntly could not have arrived at it before the eleventh 
 day. The deseent therefore from the summit of the 
 pass could not have commenced later than on the ele- 
 venth day. But Hannibal arrived on the sununit of the 
 pass on the ninth day, and encamped there for two days. 
 JSTow the eleventh day cannot be included as one of these 
 two days, for that day is the latest that can be assigned 
 to the conunenccment of the descent. The ninth °and 
 tenth days were thus the two days spent on the summit of 
 the pass : and the summit must therefore be supposed to 
 have been gained on tlie morning of the ninth day ; for 
 that day could not be included as one of the two during 
 which the army remained encamped, unless the greater 
 part of it was spent by the army in its encampment. A 
 similar conclusion must be drawn with respect to the three 
 days lost on the descent. The army arrived at the broken 
 path on the eleventh day, the first day of the descent, 
 and left it on the thirteenth. The eleventh, the twelfth,' 
 and the thirteenth days are tluis the days alluded to by 
 Polybius in the words, e. mepai, n-ptal KaKoiraO/jcjau (Chap. 
 
 c)5), which clearly refer to the hardships endured while the 
 
 army was obliged to halt in consequence of the path 
 
 being broken up. Now, in order that these tlirec days 
 
 may be referred to as passed while the army halted the 
 
 whole or the greater part of each day must have been 
 
 consumed at tlie broken patli. The army must, in conse- 
 
 quence, have had its progress arrested on the morning of 
 
 the eleventh day. The greater part of that day, the whole 
 
 of the twelfth day, and the greater i)art of the thirteenth 
 
 were spent at the broken patli. On the cvenino- of the 
 
 thirteenth day the elcpliants were brought down by tlie 
 
 road which Hannibal had restored, and rejoined that part 
 
 of the army which had previously passed. On the fifteenth 
 
 day the whole army arrived at the commencement of the 
 
 \i)lains of Italy. 
 
 ^ As Hannibal is stated to have encamped for two days 
 at the summit of the pass, the following condition at once 
 results : 
 
 ^ 
 
 
 •. T^H 'i iw . 
 
 r ■ ;■ ' ^iwm. * wm'''" j ' lim m 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 35 
 
 tU.' 
 
 *A place Ruitable for the cncanipmrnt of Ilannlbal'B 
 army must be found at the summit of the pass'.' 
 
 Also, since the descent began on the eleventh day, and 
 terminated on the fifteenth, no more than five days were 
 spent on the descent. But of these, more than two were 
 lost at the broken path: (viz.) all the twelfth, and the 
 greater part of the eleventh and thirteenth days. There 
 were therefore less than three days of actual marching 
 between the summit of the i)ass and the commencement 
 of the i)lains of Italy. This circumstance su])plies another 
 condition*. 
 
 .. 'The commencement of the plains of Italy must be 
 less than three days march from the summit of the pass.' 
 
 For several days, after leaving the town of the Allo-'^ 
 broges, the march of Hannibal was undisturbed by hostili- 
 ties : but on the fourth day he again had to encounter 
 great dangers. These dangers resulted from a treacherous 
 attack made upon the Carthaginians by the Alpine Gauls, 
 with whom Hannibal had previously entered into a treaty. 
 The attack took place near a certain strong white rock, 
 TTcpi Ti XevKOTTCTpov oyvpovy and was made on the fourth 
 day's march from the town. A di/rcrent date is, however, 
 generally assigned to this event. Jt is usually supposed, 
 not that the attack took place, but that Hannibal met and 
 entered into a treaty with the Gauls who afterwards made 
 the attack, on the fourth day. This view of the casc/^ 
 would defer the attack till the seventh day, as the Cartha- 
 ginians had marched for two days previously under the 
 guidance of the Gauls; or, at any rate, till the sixth day, 
 if the fourth day be included as one of the two which 
 preceded the attack, and during which Hannibal was 
 guided by the Gauls. It will consequently be necessary 
 to consider here, somewhat at length, what is the right 
 interpretation of Polybius' account of these occurrences. 
 
 The diirerence of the two views arises from this cause ; 
 
 1 Condition V. 
 
 2 Condition VJL 
 
 3 o 
 
 H 
 
 ::f! 
 
36 
 
 11ANNIBAL\S PASSAGE OF TIIE ALPS. 
 
 1 1* - 1 
 
 that, \Ylnlc wc have taken Polybius* nnrrative in tliis place 
 to consist, lirst of a sunnnary statement of the events of 
 four days, and then of an cxphmation and a detailed 
 account of those events : yet it has, on the other hand, 
 been generally supposed, that the whole is one continuous 
 narrative; or, at all events, that no part of the details of 
 the transactions ^vith the Alpine Gauls refers to the three 
 days preceding the fourth day indicated by the word 
 TCTapTaloS' 
 
 Now the few lines, printed in the translation of Toly- 
 bius as the sunnnary, comprise two facts ; (1) An undis- 
 turbed march for some time, i. e. three days ; and (2) a 
 great danger encountered on the fourth day. The two 
 questions thus suggested for consideration are: (1) How 
 did it happen that Hannibars march was undisturbed for 
 three days ? and (2) What were tlie particulars of the 
 danger encountered on the fourth day? These two (lues- 
 tions, according to our view, Polybius proceeds at once to 
 resolve; and his explanation either implies or states the 
 following facts. IlannibaPs march was undisturbed for 
 three days, because the Gauls thought their chance of 
 destroying the Carthaginian army would be greater, if 
 they refrained from attacking it until it arrived at a cer- 
 tain difficult pass ; and if they were enabled, by previously 
 gaining the confidence of Hannibal, and becoming his 
 escort, to assail him when he was off his guard, and%)en 
 to great danger from a treacherous attack. They would 
 besides, by these means, secure a great part of their coun- 
 try from being ravaged by an hostile army, and would also 
 have time to collect their forces together, and hang upon 
 the rear of the Carthaginians, the most vulnerable extre- 
 mity of an army upon its march. Acting in accordance 
 with this design, they met Hannibal, cither on the evening 
 of the first day, or on the morning of the second day, of 
 his march from the town ; and bore with them boughs and 
 crowns in token of their amicable intentions. Hannibal 
 liowever doubted the sincerify of their professions, but 
 tliought it at last expedient to seem to confide in them. 
 For two days he pursued his march under their guidance! 
 aheso two days were the second and third. On the next 
 
 HANNIBAL'S PASSA(JE OF THE ALPS. 
 
 37 
 
 day he was attacked by the (Janls. 'J'liis ilny was the 
 fourth : and tlie treacheroug attack was the great danger, 
 which Polybius particidarly mentions as having occurred 
 on that da3\ 
 
 Now from this view of the case it may certainly be 
 said, that the history comes out with clearness and i)rcci- 
 sion. AVhcther what has been regarded as the explanation 
 of a previous summary, were intended as such or not, yet 
 it undoubtedly does explain it j)erfectly, both with res])ect 
 to time and circumstances. Such a coincidence, it may 
 be objected, is accidental. This is possible, though not 
 highly probable ; for the coincidence is almost too exact 
 to be the effect of chance. There are, besides, other rea- 
 sons which lead us to conclude, that Polybius' history may 
 in this place be divided into a summar}, and a subsequent 
 explanation and detailed narrative. The justice of this 
 conclusion may be perceived, by comparing what is given 
 in this place as a summary, with the two passages given in 
 the translation of Polybius as the previous summaries. 
 These then are the three passages in question : 
 
 (1) ILannibal, having continued his march for four 
 successive days from the passage (of the llhone), came to 
 the district called the Island, a populous and corn-pro- 
 ducing tract of country, which derives its name from the 
 very circumstance (of its formation). 
 
 For the Phone Sec, 
 
 (2) Hannibal, in the course of ten days, having ad- 
 vanced along the river side for a distance of 800 stadia, 
 commenced the ascent of the Alps, and found himself in a 
 situation of the greatest danger. 
 
 For as long as &c. 
 
 (3) During the days innnediately following, he led on 
 the army safely for some distance : but, on tlie fourth da^', 
 he again had to encounter great dangers. 
 
 For tlie pcoj)le <5s:c. 
 
 There are here i)resentcd, in one view, what arc re- 
 garded as being three brief statements of marches, all 
 broken off at points, where observations, explanations, or 
 details, are requisite ; the unfinished sentences, which 
 
 liatHl 
 
 i*i 
 
 P* M 
 
58 
 
 ILVNNIDAL'S PASSACJE OF THE ALPS. 
 
 IIANNIDAL'S PASSACJE OF THE ALPS. 
 
 39 
 
 ti 
 
 , ! 
 
 K' ' 
 
 !i 
 
 I 
 
 
 ; :i 
 
 j 
 
 follow in each case, being considered as the beginnings 
 of such particular remarks. Now, let the events which 
 occurred, between the passage of the Ehone and the en- 
 campment at the town, be considered. They will be found 
 to comprise these things ; a march of four days along the 
 Ehone to the frontier of the Island, a num])er of episodical 
 transactions at the Island itself, a march of ten days to a 
 certain town, not flir from the connnencement of the ascent 
 of the Alps, and of a battle near the town and its subse- 
 quent capture. In the first two supposed summaries, these 
 two marches are given, the Island noticed, and the dangers 
 mentioned which bcfel the army at the commencement of 
 the ascent of the Alps. All the leading events therefore, 
 which happened between the passage of the Ehone and the 
 encampment in the neighbourhood of the town, are set 
 brielly before the reader. This being the case, these sen- 
 tences may reasonably be looked upon as sunmiaries of 
 these important occurrences. It remains to be considered, 
 how far, in these two instances, the remainder of the nar- 
 rative is explanatory of these short statements. 
 
 Now the first summary presents to our notice the 
 march to the Island, and the Island itself Of the march 
 itself, nothing is subsequently said. It ofFered, we may 
 suppose, no circumstances, which called for any particular 
 observation. All the latter part of the forty-ninth Chapter 
 is devoted to a description of the Island; to an account of 
 the events which took place there, and which formed a 
 kind of episode to tlie great expedition; and to a pro- 
 spective glance at the assistance, which the men of the 
 Island rendered to Hannibal during the next part of his 
 march. All these circumstances may be considered as 
 incidental to what is stated in the summary. 
 
 In the second summary are introduced, the march to 
 the commencement of the ascent of the Alps, and the 
 dangers which the Carthaginians encountered in that 
 neighbourhood. The long account that follows explains 
 why, during that march, the army was unassailed by the 
 Allobroges, gives full particulars of the dangers it after- 
 wards passed through, and brings it to its encampment at 
 
 the town of the Allobroges. Tlicse events also are all suIj- 
 ordinate to those in the previous summary, and carry the 
 history no further onward than where the dangers it alludes 
 to terminated. 
 
 Let now the three passages given in page 'M, be com- 
 pared together. The third will, without much difliculty, 
 be seen to have a considerable similarity to the other two; 
 especially to the second, which it resembles with respect 
 to the events it records. In both instances, there is an 
 undisturbed march, succeeded by a dangerous contest. 
 As such is the case, it is not unreasonable to suppose, that 
 the historian would treat the last of these sectioris of his 
 history in the same manner as the previous section : that 
 having (as he unquestionably does in the previous instance) 
 brought plainly forward in a short statement, the march 
 itself and the dangers at its close, h(i would then j>rocecd 
 to enter into the requisite details, as he had done just be- 
 fore in that parallel case. This, it will be seen, from refer- 
 ence to page 37 is exactly the mode of narration which it 
 has been concluded that he adopted : and it has already 
 been noticed, how clearly his narrative developes itself on 
 such a supposition. The views, therefore, which have here 
 been taken, with respect to the events which occurred on 
 the fourth day after Hannibal left the town, seem to be 
 perfectly in accordance with Polybius' mode of narration, 
 besides bringing out the history with great clearness, and 
 explaining everything simply and without confusion. 
 
 The generally received interpretation of Polybius' 
 narrative is nmch less satisfactory. According to this 
 interpretation, Hannibal met the deimtation of the Gauls 
 on the fourth day after he left the town, and must conse- 
 quently have been attacked on the sixth or seventh day. 
 Yet Polybius states, (fixing the date with careful precision, 
 fi^tj Si) T€TapTatos wv), that Hannibal fell again, tvlien now on 
 his fourth dcvjs march, into some great danger. What then 
 was this great danger? According to the view we have 
 taken, it was the treacherous attack in the neighbourhood 
 of the rock : according to the general view, it was the 
 meeting with the deputation of Gauls bearing boughs and 
 crowns. But this last supposition is scarcely well founded : 
 
 :■ 
 
 r 
 
 
 tr- • 
 
 '^ 
 
( 
 
 40 
 
 -i ' 
 
 1 i 
 
 ■4 
 
 I 
 
 Li 
 
 iil 
 
 I! 
 
 ■u 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 for no nrvcat danger could l)o s.-iul to result to iramubiil 
 IVom such an encounter, or from such weapons. Xcither 
 is this difficulty by any means explained away by the fact, 
 that the intentions of these Gauls were treacherous, and 
 that, two or three days afterwards, they brought Hannibal 
 into great danger. It was the danger itself, and not the 
 Gauls from whom he afterwards incurred that danger, that 
 Hannibal encountered on the fourth day. Nor is it allow- 
 able thus to compress the events of three or four days 
 into one ; as must be done, if the dangers of the attack 
 at the Eock, on the sixth or seventh day, are supposed to 
 be comprised in the pacific interview of the fourth day. 
 The evident intention of the historian to fix his dates with 
 accuracy must be completely defeated by such a lax inter- 
 pretation. Far preferable to a strained explanation of this 
 nature must be the one, which takes the great danger 
 spoken of to be nothing else than tlie treacherous attack 
 of which Polybius afterwards gives the details, a time 
 of peril when tlie whole army was on the very brink 
 of destruction. This was, in absolute reality, a o-reat 
 danger: the meeting with the Gallic deputation could 
 only, at the most, be considered as such in pro.si)ect; 
 being, in fact, only partially conducive to the great dan- 
 gers, which bcfel the Carthaginians several days subse- 
 quently. 
 
 Yet Polybius docs not merely say, that IIanni])al had, 
 on the fourtli day, to encounter great dangers ; but that he 
 had af/aliiy on the fourth day, to encounter great dangers. 
 To what previous event then does this *ar/aui' rclcr? 
 Clearly to the similar part of what has been given above 
 as the second summary, where, in page 37, it is stated, 
 that Hannibal found himself in a situation of the greatest 
 danger. This danger, we know, bcfel him in consequence 
 of the attack made upon him at the commencement of the 
 ascent of the Alps : and the almost complete identity of 
 the two statements of danger would lead us to expect 
 something similar on the second occasion. The attack 
 near the Pock satisfies these expectations completely : but 
 no resemblance can be found between the battle with the 
 Allobroges, and Uic peaceful conference with the Gauls, 
 
 t: ■ 
 
 m- 
 
 y 
 
 1: 
 
 
 A- 
 
 HANNIBAL'S PASSA(JE OF THE ALPS. 
 
 41 
 
 who bore boughs and crowns in token of nmity. Yet this 
 is the only event which took place on the fourth day, if 
 the general interpretation of Polybius be adopted. 
 
 Lastly, it may be urged, that when Polybius says, *but, 
 on the fourth day he again had to encounter great dangers,' 
 the historian plainly wishes to mark the precise date of 
 some important occurrence; in like manner as he after- 
 wards fixes the great event of the arrival at the summit of 
 the pass, as taking place on the ninth day. J3ut which of 
 the two events was pre-eminent in point of importance : 
 the attack by which the whole Carthaginian army was 
 nearly brought to destruction, or the meeting with a depu- 
 tation of Gauls bearing boughs and crowns? There can 
 evidently be no question as to the relative importance of 
 these two events : nor, as has been before observed, can 
 the connexion of the first with the second make it be con- 
 sidered as identical with it in rcsjjcct of date. 
 
 From all these arguments it appears, that the inter- 
 pretation of Polybius, generally accepted, is found, on 
 close examination, to be without foundation; the only 
 satisfactory view that can be taken of the Greek narra- 
 tive being, that Hannibal was attacked, near * the strong 
 white rock,* on the fourth day of his march from the town 
 of the Allobroges. In the investigation of Hannibal's 
 route, it is exceedingly important to determine the date 
 of this attack with accuracy: for the 'strong white rock' 
 seems noticed by Polybius, expressly with the view of 
 pointing out the locality where the contest occurred, by 
 mentioning a striking and permanent landmark, which 
 offered itself in that place as a natural monument of the 
 battle fought around it. The place where this remarkable 
 rock, distinguished by its strength and whiteness, is to be 
 sought, must be found by fixing the date of the events 
 which occurred in its neighbourhood; and these events, 
 we have now concluded, occurred on the fourth day after 
 Hannibal left the town. As, therefore, Hannibal remained 
 encamped near the Pock durhig the fourth night, and as 
 he arrived at the summit of the pass on the ninth morning, 
 the Pock must be situated nearly half-way between the 
 town and the summit of the pass; nearly half-way, that is 
 
 » f \ 
 
 I 
 
 
 ir 
 
 > 11 
 
 m 
 
 In I 
 
 I ft ] 
 
 \ti H 
 
 !-"«■.'•«•' -js':,-*- rjarit-T-'™-*!,' 
 
y 
 
 42 
 
 HANNIBAL'S PASSxVGE OF TIIK AM\S. 
 
 to say, in point of time, for, in point of distance, the 
 respective diflicultics of the way, above and below the 
 rock, must be taken into account. Another condition for 
 the determination of Ilannibars route is thus obtained*: 
 
 * At a place nearly half-way, in point of time, between 
 the town of the Allobrogcs and the summit of the pass, a 
 certain ' strong white rock' must be found ; and the adjacent 
 country nuist be in accordance with the events, which are 
 recorded by Polybius to have taken pkee in the neigh- 
 bourhood of the rock.' 
 
 Wc have now to consider Polybius' account of the 
 battle which was fought near the rock he so particularly 
 mentions. The Carthaginian army was attacked, he says, 
 while it was passing through a dinicult and precipitous 
 ravine, (pdpayyu Tiva Suaftarou Kal Kptj/xi/wctj. By this 
 . expression he probably means a narrow passage bounded 
 by precipitous acclivities, where the way presented con- 
 siderable difficulty. Polybius afterwards calls this ravine 
 Xc^pdSpa, and says that the army e^e/xyjpiaaTOy dejiled 
 through it; an operation which occupied a whole night. 
 Prom this also w^e should be led to conclude, that there 
 was a narrow passage, where not many could i)ass at a 
 time, through which the Carthaginians were obliged to 
 make their way. This c^dpayl^ or x'^P^k'^ ^viU form 
 one of the features of the country, which must be sought 
 for in the neighbourhood of the * strong white rock.' 
 
 As Hannibal proceeded through the country of these 
 Alpine Gauls, the band, which had first met him on his 
 march from the Town, led the way in the van. But as the 
 Carthaginians advanced, other bodies of the Gauls col- 
 Iccted on the rear. Each village or district that was 
 passed through, would send out its men to swell the num- 
 bers of those who were following up the army; so that, 
 in the two days, during which they allowed Hannibal to 
 march peaceably through their country, they would be 
 
 Condition IV. 
 
 
 HANNIBAL'S PASSACJE OF THE ALPS. 
 
 43 
 
 enabled to draw together a considerable force. 'Jliese 
 movements increased the suspicions of the Carthaginian- 
 general, which had never entirely been laid asleep. Aware 
 of the danger to which liis rcnr wms cxposcid, \n\ placed 
 his heavy infantry there, to make head against the Gauls, 
 if they should venture to assail him. The cavalry and 
 baggage were at the same time i)laccd in the van, so as 
 to be beyond the reach of the (Jauls hanging upon the 
 rear. AVhat position in the line of march the light in- 
 fantry occupied is not mentioned : they were probably in 
 the centre, ready to move upon any point where their 
 assistance might be required. That they should have 
 been in a position where they could be easily brought 
 into action, was almost indispensable ; for they alone were 
 serviceable in mountain fighting, or capable of driving 
 back an enemy \\\)im didicnlt ground. 
 
 This then was the arrangement of the Carthaginian 
 line of march previous to the attack. The cavalry and 
 baggage were in the van, the light infantry probably in the 
 centre, and the heavy infantry in the rear. One body 
 of the Gauls, now about to put their treacherous design 
 into execution, was leading the way at the head of the 
 Carthaginian columns : another body was hanging on the 
 rear. The army of Hannibal was thus threatened at both 
 extremities of its line of march. 
 
 The numbers of the Carthaginians amounted at this 
 time to about 40,000 \ The line of march of an army so 
 numerous as this, especially when confined in a mountain 
 valley, must have extended to a length of some miles. The 
 foremost columns and the rear-guard would consequently 
 
 be several miles apart. 
 
 The Gauls at length made their attack: but the pru- 
 dent arrangements of Hannibal saved the Carthaginian 
 army. The attack upon the rear ndlcd ; for the heavy 
 
 1 After tlio passage of tlio Hlionc they amounted to 4G,000, of whom^ 
 26,000 reached tlio plains of Italy. Nearly as many were lost upon the 
 descent from the Alps, as there wore during the ascent. About 3r,,000 
 would thus reach the summit of the Alps. Their numbers at the battld 
 of the Rock may therefore bo taken to have been about 40,000. * / 
 
 
 i 
 
 \\ 
 
 
 ■Ww^iSfJiB 
 
u 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 45 
 
 infantry succeeded in willistandins,^ (lie cdbrts of the enemy. 
 The contest here probably took i)iace in a part of the valley 
 ^vherc the ground uas not difficult ; for the heavy inHintry 
 >vould hardly have repelled the uiountaineers, if Ihcy Iiad 
 had at the same time to contend uith the ru^^-n-edness of 
 the mountains. Yet, though the CarthaginianTAvere suc- 
 cessful upop this quarter, they suffered severe loss, both 
 in men, horses, and baggage-animals. The van therefore 
 was also attacked, for it was there that the cavalry and 
 baggage had been placed by Hannibal. This attack must 
 have been made by the Gauls who were previously men- 
 tioned by Tolybius as leading the way in the van. They 
 were, however, probably not alone, but had been joined by 
 other parties of Gauls, who were awaiting them there; 
 yet of whose presence Hannibal does not seem to have 
 been aware, as it was his rear which he appears to have 
 thought especially in danger. It is not likely that any 
 of the Gauls who attacked the rear had any share in the 
 destruction of the cavalry and baggnge-animals in the 
 van : for they could hardly have managed to make their 
 way along the mountain sides, unopposed by Hannibal, 
 \ for so long a distance. Neither could the heavy infantry 
 be spoken of by Polybius as rendering the service they 
 did to the army, had the Gauls who had hung upon the 
 rear succeeded in passing all along the mountains so as 
 to fall on the van. AVhen the historian says, toutwp Si 
 eCpeSpeuovroiv eXarrov auvefi)} y^'eaOai rj irdOos' ovtoi yap 
 f-are^av ti]v eiTLCpopuv twv fiapfidpwvy it seems necessarily 
 to be implied, that the heavy infantry, repelling the enemy 
 in the rear, succeeded in preventing them from iuflictino- 
 any injury upon the Carthaginians. 
 
 The Gauls who attacked the van had the advantage of 
 the ground. As they marched in advance of the Ca'rtha- 
 ginians, they were enabled to occupy the slopes above the 
 road, along which the army which followed them had to 
 proceed. From this position, (joined perhaps, as has been 
 conjectured, by others of their own countrymen) they 
 assailed the Carthaginian van with rocks and stones. Down 
 the steeper declivities they probably rolled rocks from 
 some distance above : and, where the slope was not suffu 
 
 fl 
 
 
 
 
 !<•' 
 
 
 cicntly steep to give the rccpiisite inipulse to the rocks, 
 they advanced almost close to the Carthaginians, and 
 hurled heavy stones at them as they passed beneath. ]^y 
 these attacks a heavy loss in men, horses, and baggage- 
 animals, was sustained. The foremost column was, in all 
 probability, cntircl}^ destroyed, before Hannibal could take 
 any measures for the defence of this division of his army. 
 Seeing the destruction that awaited them, the rest of the 
 cavalry, with the baggage, would halt or fall back, until 
 a safer progress could be ensured for them. y^ 
 
 No danger was now to be aj)prehended on the rear*"^^ 
 the heavy infantry there held the Gauls in check, and 
 Hannibal was enabled to devote his ])ersonal efforts to 
 the safety of the van. For this ])urpose it must have been 
 necessary to gain i)ossession of the heights above the 
 slopes, where the Carthaginians had suffered so severely 
 from rocks and stones. The light infantry would be the 
 arm to which this service would naturally be entrusted ; 
 but it cannot be said whether they composed all the 
 forces on the occasion. One half of the Carthaginian 
 army, that is to say, about 20,000 men, were led on by 
 Hannibal in ])crson against the Gauls on the mountains, 
 and succeeded either in driving them back, or in manoeuv- 
 ring so as to make them abandon their posts. However 
 his purpose may have been effected, he at length took up 
 a position where he was able to guard the rest of his army 
 as it resumed the interrupted march through the ravine. 
 This march was performed during the night, which may 
 have been about to fall when Hannil)al took up his posi- 
 tion on the heights. He probably thought that during the 
 night he could draw his army off better from the Gauls in 
 the rear, who ])erhaps would not susj)ect that such a 
 movement was contemjdated, even if the fatigue and ill 
 success of the conflict of the day had not rendered them 
 indisposed for an innnediate renewal of hostilities. During 
 all the night Hannibal remained in ])osition, separated 
 from the rest of his army as it defiled through the ravine. 
 On the morrow, the Gauls having withdrawn, Hannibal 
 rejoined his cavalry and baggage, and resumed his joinncy 
 towards the summit of the Alps. ^ 
 
 M 
 
 
 '•I 1 
 
 i\ 
 
 m 
 
 H 
 
 Til 
 -I 
 
 4 
 
 fl 
 
 'f-^l 
 
4G 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 Such, as far as can be collected from the narrative of 
 Polybius, were the progress and termination of the battle 
 of the Hock. The most remarkable circumstance it con- 
 tains, — a circumstance which gives an important clue by 
 
 which the scene of this contest may be found, is the fact 
 
 of Hannibars having posted 20,000 men on the heights, 
 
 away fi-om the rest of his army, and for the sake of en* 
 
 suring its safety. This circumstance at once suggests the 
 
 existence of an extent of practicable ground above the 
 
 slopes on one side of the road, by no means usually to be 
 
 found in the Alps. It is also worthy of notice, that 
 
 Hannibal is said to have passed the night ^cofyk tcZi^ Ittttwu 
 
 Kal Tiov vTToX^vyi(i)Vy an expression which seems to imply 
 
 that his position was at some considerable distance from 
 
 the place where he had left his cavalry and baggage; - 
 
 for a sliglit distance, such as some fifty or a hundred 
 
 yards, between two divisions of an army, would hardly be 
 
 considered as amounting to a separation of positions, or 
 
 Iiave been a sufriciently important movement to have been 
 
 recorded. 
 
 r>nt tlie extent of i\\c practicable ground on one side 
 of the Carthaginian army seems to have been more than 
 was merely sufficient for the posts of Hannibal's 20,000 
 men. For these men were placed on the heiglits in order 
 that they nught protect the rest of the army, from which 
 they had separated. They had consequently,— at least 
 such a conclusion api)cars necessary,— interposed between 
 tlic Gauls on tlie heights and their own cavalrv. These 
 Gauls would thus be stationed, in all i)robabifity, on a 
 more remote part of the same tract of ground where Han- 
 nibal had posted his 20,000 men. 
 
 The difficult and precipitous ravine, mentioned by 
 Tolybms, has been already noticed, and its character con- 
 jectured. Tlie slopes where the Gauls attacked the flank 
 of the Carthaginian van, by rolling down rocks and castin- 
 stones ; and the great extent of practicable ground above 
 these slopes, are the chief features (besides the 'strono^ 
 ^vhite rock') by which the scene of the battle is 'to be 
 recognized. We have therefore to fmd in the immediate 
 neighbourhood of the Jlock : 
 
 %. 
 
 % 
 
 % 
 
 ^ 
 
 HANNH^AL'S PASSAGE OF THE ALPS. 47 
 
 (1) A difficult and precipitous ravine, through which 
 the road must pass. 
 
 (2) A range of slopes overlooking the road, and liable 
 to be swept, in one or more places, by rocks set in mo- 
 tion from above. 
 
 (3) A large extent of practicable ground above these 
 slopes. 
 
 (4) A position on this ground, at some distance from 
 the road, where 20,000 men could be stationed, so as to 
 protect completely an army marching along the road be- 
 neath, and such as would not expose themselves to be 
 attacked at any serious disadvantage, by an enemy lying 
 beyond them relatively to the road'. / 
 
 The portion of Polybius* narrative, relating to the\ 
 march from the neighbourhood of the ]iock to the summit 
 of the pass, is very short, and presents nothing of much 
 consequence. It comprises a short statement of the marcli; 
 and mentions how the Carthaginians were harassed on 
 their way by occasional jircdatory attacks on the ])art of 
 the barbarians, but had not to encounter any formidable 
 opposition to their progress. Tlic only particular obser- 
 vation, which Tolybius makes upon this ])art of the march, 
 refers to the cfrect which the sight of the elephants pro- 
 duced upon tlie barbarians ; who were so overawed by the 
 appearance of these unknown and gigantic animals, that 
 they were deterred from making any assault upon those 
 parts of the line of march where they were to be found. > 
 It does not appear, that these barbarians formed any part 
 of those who had attacked Hannibal in the neighbour- 
 hood of the Iiock, and wlio would naturally have been left 
 behind by the Carthaginians. Indeed, had the barbarians, 
 who made the jn-cdatory attacks, had any share in the 
 battle of the Hock, the appearance of the elephants could 
 
 ^ As no attack is said to have been made, citlicr on tlic army during 
 itsniglit march, or on tlic protecting troops of Hannibal, we may conclude 
 that neither division was much exposed to molestation on the part of tho 
 enemy. 
 
 
 r : 
 
 i.; 
 
 
 
 r 
 
 11 
 
 *- il 
 
 y 
 
48 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 
 hardly have been so iinfaniiliar lo thcin, or have overawed 
 them in the manner it is said to liavc done. JNIost pro- 
 bably, they were merely the inhabitants of the several 
 districts through Avliich the army successively passed, who 
 seized any favourable opportunity of plundering that oc- 
 curred, without offering any organized resistance to the 
 Carthaginians. 
 
 It is, however, Jiecessary to notice in this place an 
 erroneous interi)rctation of a passage in Tolybius, from 
 which it has been concluded, that the battle of the Eock 
 took place on the day before Hannibal reached the summit 
 of the Alps. The Greek narrative runs thus : rij ^' cTraJ- 
 piov Twu TToXeixicov -y^wpiaOeuTcoi', auuu\j/a9 toIs 'nnrevai koI 
 
 Twu AXwewv, /C.T. X. 
 
 In this passage, it has been supposed that the words 
 TJ5 ^* iiraufnou arc connected with TrpnPjye, and that Han- 
 nibal consequently gained the summit of the pass on the 
 day after he fought the battle. Yet this supposition is 
 unfounded, for all that the Greek imphes as having oc- 
 curred 'on the morrow,' is the junction of the two^divi- 
 sions of the army after the enemy had withdrawn. It is, 
 besides, by no means natural to suppose, that the day 
 mentioned as the eiraipiov is the same day as that indi- 
 cated subsequently in the word ewaralos ; for the latter day 
 seems clearly to be spoken of by Polybius as later than 
 the former. Neither would any sunicicnt time be left, if 
 this view were adopted, for the scries of repeated attacks, 
 on various parts of (lie army, and in diilbrcnt localities, 
 which are recorded to have been made between the neigh- 
 bourhood of the Hock and the summit of the pass. For 
 Hannibal reached the summit of the pass on the morning 
 of the ninth day ; so that scarcely any time could inter^ 
 vene between his arrival there, and his departure from his 
 position near the Hock, if it were only on the same ninth 
 morning that he left this position. Indeed, it seems plain 
 that the Eock must have been at a distance of many miles 
 from the summit of the pass, for the predatory attacks, 
 which the Carthaginians suffered during this part of the 
 march, were, in all probability, made by the inhabitants 
 
 
 >H^t'' 
 
 ^ 
 
 ij 
 
 
 1 
 
 I 
 
 HANXIDAL'S PASSAGE OF THE ALPS. 
 
 49 
 
 of the several districts through which they passed^; and 
 the immediate neighbourhood of the summits of the 
 Alpine passes was not inhabited in the time of Polybius, 
 as indeed is generally the case at present, although the 
 uninhabited regions are now% no doubt, less extensive. If 
 the words also of Polybius — oXoax^pf^l MeV ov^€v\ irepiiriiT' 
 TWU €Ti avarijfiaTi twv (iapftupwi/y /card fjtcpt] Sc kui Kara 
 
 T07rOV9 TTdpCVOyXoVILCVO^ VW UVTtOV* IVV o\ /<tr UTTO T>/S OVnU' 
 I * ^\ * \ -* /•/ ^ /' 
 
 7fa9 ot ce ttTTO T^^ irpcoTOTTopcia^ uTreaircou rwv (TKevocbvptov 
 €via, TrpoairiTTTouTe? evKaipcos* k,t.X. — if these words be 
 duly considered, it will be seen, that the events to which 
 they refer require for their occurrence a considerable 
 space of time and length of march, and that they are not 
 such as would have happened in the course of two or three 
 hours, or on a march of a very few miles, especially 
 through a district where there were scarcely any, or no 
 inhabitants. That the 'strong white rock' of Polybius 
 should thus have been situated (as is the case with the 
 *Eoche Blanche' on the road of the Little St Bernard, 
 and the gy])sum rock near Termignon on the road of the 
 Mont Cenis) at a distance of no more than about six 
 miles below the summit of the pass, appears, from this 
 passage alone of Polybius, to be a supposition deficient in 
 probability, even if no account be taken of the jn'cvious 
 part of the narrative, which relates to Hannibars transac- 
 tions with the barbarians who afterwards betrayed him, 
 and from which the natural conclusion is, that it was on 
 the fourth day that the Carthaginians were attacked near 
 the ' strong white rock.' Upon tliis latter supposition, on 
 the other hand, ample space and time are allowed for the 
 events which took place between the Jiock and the summit 
 of the pass; Hannibal being suj)poscd to have left liis 
 position near the Jlock on the morning of the fifth day, 
 and to have arrived at the summit of the pass four days 
 afterwards. 
 
 ^ As the van of Hannibars army was plundered, as well as tho rear, it 
 Bccms almost certain that tlicse attacks could not have been made by bands 
 following up tlie Carthaginians. 
 
 it 
 
 Ij 
 
 i 
 
 
 (f a 1 
 
 ii 
 
 m 
 
 
 .^• 
 
 4i «■ 
 
 '}-■ 
 
 if 
 
 f^ 
 
 i\ 
 

 \i 
 
 u 
 
 I : 
 
 CHAPTER V. 
 
 Discussion of Poljbius' namtivo of tl.o events «I,ich took place at tlio 
 summ, of tl.o pass and on the descent to Turi„._The dah a o 
 
 tlio descent.— 11,0 path destroyed by a landslip.— Ma.u.er in which 
 U.0 path was .econstructed.-Hannibal arrives in the connt J of t ! 
 ramnn.-Journal of tho march fron> the confluenco of tl o Rhon 
 and Isuro to the conimonccment of tho plains of Ital v f 1 r • 
 for the detennination of tho pass which m"„ibalcross;:^'"" 
 
 0''townTft"M';""'""^ '^'''^"^'^^ ''« ^'^'^ left the 
 y to^vn of the Allobrogcs, Hannibal at Icn-th rr-iincd 
 the summit of the iviss Tr«,.„ i.„ . '^-I'ocn ganied 
 
 ^, '•"'- l^'^ss. llcic lie encamped, and rcnninod 
 
 dunng the greater part of the nintit ind all the eS 
 day, waitmg for stragglers ^vho had been left behiml and 
 
 oTuS TeirT '^ '-T 'T ^"^^ *''° *«"^ ^^^^ 
 
 hostile Xcl s it '^'\ . T '°" ""'^ '^ ^''"-^^^^ fro-n 
 nosuie attacks they indeed obtained, but were exnoscd 
 
 n return to new and equally depressing hardshS it 
 .^s now the end of Oetober. Winler had^already 'o gned 
 for a month upon the higher parts of the Alpine mfscs 
 
 longer sustained them • nnd 1,, I.^ i '''"'C'dty no 
 
 their thoughts naturallVre^eied ^U^IS^ t^^l!-;, 
 they had only cseaped with diflieulty and c^Vv Toss "n 
 a renewal of which they expected to'^awrdt heh^ en e^b^d 
 ^id reduced numbers upon the descent. l^ci movf 
 sions also -were DroI)n]Jv u.r +i • i- -Lucir pioM- 
 
 haustion. anrth^ey Sd ^n ,""° "''P'-^--^^'""? to e.x- 
 
 snowy crest of tloA in -M ?''''■'=' encamped on the 
 
 for n^i:::rom,aThe1^ He e^ort^ l'"';':-'° ^•^'^"^""»" 
 or the spirits of the army si^lSur AV;irrf\Spr 
 
 :.r 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 51 
 
 cncy spread tliroii^rli i],c raiilca of tlio Cartlm^Mniaim : and 
 on the tenth day ', ailer they liad been exposed to the 
 severities of a uinter^s night upon the Alps, their profound 
 dejeetion aroused the anxiety of their leader, ^vlio imme- 
 diately took measures to counteract a state of feelinrr, which 
 might have proved fatal to the success of the expedition. / 
 From this circumstance we are made aware of one"^ 
 of the most important characteristics of the pass which 
 Hannibal crossed. He Iiad, Polybius says, one only re- 
 source by which he could encourage his desponding army: 
 and that one resource consisted in setting before them 
 the prospect of Italy. Italy was thus visible from the 
 summit of the pass. No modified interpretation^ of this 
 assertion can be admitted : it cannot be supposed that 
 Polybius would have spoken of Hannibal, either as shewing 
 or as pointing to, that which was at the time out of sight. 
 Nor 13 It easy to conceive, unless Jtaly itself hiy before 
 their eyes, how the soldiers would have derived much en- 
 couragement from all Hannibal's words. For this it was 
 essential that Italy should be visibly present, and their 
 extrication from tlie fatal region of the Alps evidently at 
 hand, the utmost limits of the mountains, and the country 
 beyond, being plainly descried. Otherwise Hannibal would 
 have had no foundation on which to ground his encourag- 
 ing exhortations; but all his words would have fallen 
 dead, had the Carthaginians seen, as they would ha^'p 
 done on most passes, that the mountains still shut them 
 m on every side, and that ' Alps on Alps' still lay piled 
 before them upon their future course, all the unknown 
 and intricate vallies and ravines among them having still 
 to be threaded, before they could hoj)c to arrive at the 
 invisible i)lains of Italy. y 
 
 Yet, though Italy was visible from the summit of thc^ 
 pass, it was not from the i)lace where the Carthaginians 
 were actually encamped, that Italy could be seen. On 
 
 1 It was on the tenth day ; for on tlie morrow tlicy began to descend, 
 and their descent commenced on the eleventli day. 
 
 2 Sec Deluc, Ilistoirc du jmssagc dcs AIpcs, pj). l.'iG— ir,0; nnd Dhscr^ 
 talwn^on the j.assar/e of Hannibal, by a member of (he University of Oxford, 
 
 4—2 
 
 
 N * '! 
 
 VI 
 
 
 I 
 
 tl 
 
 
 
 5 
 
52 
 
 HANNIBAL'S TASSAGE OF THE ALPS. 
 
 'J 
 
 the tenth morning, after they had remained encamped for 
 a whole day, the ivdpycia t7i^ 'IraX/ac, or 'manifestation of 
 Italy,' had not taken place. From their encampment then, 
 at that time, Italy was invisible ; and Hannibal must have 
 brought them to some other point, from whence the pro- 
 spect could be obtained. As to the position of this point 
 of view, Polybius says nothing; but we are led to infer 
 that it lay not far distant from the road over tlic summit 
 of the pass. For, had it lain at any great distance, it is 
 rather di/liculfc to imagine how Hannibal had become 
 aware tliat there was any point commanding such a view; 
 a circumstance, of which he was probably apprised by his 
 Gallic guides, avIio would only have been accpiaintcd, from 
 previous knowledge, witli what lay in the immediate neigh- 
 bourhood of the road; as it was not likely that any one, 
 crossing these desolate regions, would have diverged un- 
 necessarily from tlie track. Besides, had Hannibal made 
 an excursion of any Icngtli for the purpose of ol^taining a 
 view of Italy, Polybius would naturally have taken some 
 notice of it, and not have passed it over in complete silence. 
 We are therefore led to conclude, that the point from 
 which Italy was seen lay not far from the road over the 
 summit of tlic pass, but yet was not included in the ground 
 where the Carthaginians encamped, when they first gained 
 X the summit. 
 
 When Hannibal shewed Italy to his troops, he pointed 
 out to them the plains of the Po. and indicated to them 
 the quarter in whicli Home itself lay, t3. t,?^ 'P.;/.,;, airTj, 
 TOTTou, ihe district described as the plains of the Po is 
 wcl known. It is the vast tract of country, almost en- 
 tirely level, which lies between the Alps on the Xorth and 
 the Apennines on tlie south, and which stretelics eastward 
 from the roots of the Cottian Alps as far as tlic Adriatic 
 sea A part of this country Hannibal pointed out to his 
 soldiers. He also indicated to them the quarter in which 
 Eomc Itself lay. Here the word is much vaguer than 
 when the plains are spoken of. In the first case it is 
 cvde,,,vv^,,^^,^ m the latter iTroSaKvicvv. Neither is it actu- 
 ^ ally Home that is spoken of, but only the quarter in which 
 Home lay. Hannibal gave his army, in fact, a vague no- 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 53 
 
 J-- 
 
 r- 
 
 tion where Home was situated. He saw, we may imagine, 
 as he would naturally do from Iho Alps, the range of the 
 Apennines rising beyond the i)lains of the Po ; and gave 
 his men some idea of the situation of Jlomc, by saying 
 that it lay beyond the range of mountains, which they then 
 descried upon the distant horizon. How far it lay beyond 
 them, he probably did not think it necessary to state, even 
 if he well knew himself; for his object was not to convey 
 information, but encouragement : and he would not liavc 
 been unwilling to let his army infer, that Home was situ- 
 ated at a short distance beyond the Apennines. With 
 respect to the j)lains of the Po, no deception was possible: 
 they could not have supposed that they saw them, unless 
 they actually did so. With respect, however, to the posi-^ 
 tion of Home, they had no means of determining whether 
 what Hannibal said was strictly true or not ; but must 
 have taken it for granted, without question, that Home 
 lay where he told them that it did, unless such an asser- 
 tion was palpably extravagant. 
 
 It is unnecessary to offer any remarks upon the saga- 
 city which Hannibal evinced in introducing the name of 
 Rome, and in connecting that city with the prospect the 
 Carthaginians then had beneath them. But a short time 
 before, seeing themselves enclosed on all sides by the 
 snows and crags of the Al])s, they had abandoned hope, 
 and given way to despair. Now, not only did the fertile 
 plains of Italy greet their sight; but it even seemed, as 
 their eyes possessed themselves of the country ui)on which 
 they looked down, and rested upon the distant Apennines, 
 which alone, as they believed, separated them from Home, 
 that the great city itself, the prize for which they had 
 dared so much, was then almost within their grasp. 
 
 It was then the sight of the Apeimincs, which had, in 
 all probability, suggested to Hannibal the idea of Home, 
 and b}' means of which he indicated its situation. The 
 condition, therefore, which results from the fact of Italy 
 having been seen from the summit of the pass, will be of 
 the following nature * : 
 
 1 Condition VL 
 
 ■1 
 
 -■•! 
 
 
 ^ yilF •r^'* 
 
5-Jt 
 
 
 ; ) 
 
 n ' 
 
 L^ii 
 
 IIA.Nx\IBAL'S PASSAGE OF THE ALPS. 
 
 the I::;;;;tytt'prb"Scr^ ['- '-r '- --^ -- 
 
 the CartLao-inians wo ul'.l . ^ " ."''°" ^''^^ ground where ] 
 the Apennines alstoulh[t?beSsibS: '" "^" P^'^'^''^''"'^' 
 
 US 
 
 "«.,.°".,!:r ss:,f";,.ty srrs t,j»™/"= -^ ^ 
 
 was excessively steon on,! ,/'."*'''» "hich their route lay, 
 tiall, eoneealo^ b^t^; ow""X'' rcV^'""^ ^^ ^^^ 
 certain where to tread missed H.of"^', ^«"S"l"C"ce, un- 
 hers, both of men ami hZT "•'^"' ^«o''"S' --^'"l great num- 
 
 cip.ccs. Of the multitudes ost iLn , ? *''° P''^" ^ 
 
 .tnd amounting to nearly 10 OOO ^ '? '''''''^ tlesccnt. 1 
 
 X probably perished h I^tI c leum't: "^"' ^^°'"^""" 
 
 -other ch...c.:c-rL.:e of tbejr:?:^^ ?--''■*• 
 crossed'. ' — -:i ^e ^-^rLHigiina 
 
 racter.' ^' ^° ^^ '"^ precipitous cha- 
 
 them S iTsSc^h W 'rr ^^-- *'- --^ ^ore 
 ealamitics. On the otW sM^ /? ^^-Tcrienccd similar 
 
 bahl, been exposed t"somoU.t'':/Z "^^ '-» P- 
 although Polybius wlm Jc ,.„ ° • "° ^^'"^ nature, 
 
 latter Jart of tl " Lcint dooT "T'"" "'"' ^^^P°^^ ^^ "»« 
 kind. Yet a stilf e ^l-ettHr '"'"'"f "/"^'"""^ '^^ ^^^^ 
 -/ginians. They arriycd at a n '^ "T ^f'^ *''« C^^^'^" 
 ran along the face of a er„ J '''"''^ ^''•="- ^'"« "^ route 
 
 path had been "m^d^Z'^^f^'^-'r- "''^ ^'''-•'^ ^^^^ 
 1-ecipiee, by a rec t laSbhr Tl"""' "" '"^ ''^ *'^« 
 
 that it was an accident of \,j.r J"" "° '^"'* 
 occurrence in tlic AIn« ^^ \"f^"^^» K very common 
 
 * Condition VIIL 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 55 
 
 both from the events which took i)lacc, and from the fact 
 j of the Carthaginians being obliged to ]>alt for three days 
 m consequence of the accident ; besides, Polybius after- 
 wards speaking of this spot, (and i)crhaps exclusively of it ) 
 uses the words twv npocp.if.^no h,„„wwv. Hannibal also 
 when he remedied the nn-sfortunc. 'built up the j.recipice ' 
 rov Kpnixvov c^owoo%*c< ; so that it is quite certain there 
 was a precipice here. ]5ut this san,c j.lacc is also called 
 aTopfm!^j and the unopfni^ is f.nther desciiijcd as recently 
 liaWov en aTreppmyvra ; in which descri])tion a slip of the 
 Bide of the precipice is evidently alluded to. The de- 
 struction also, whole or partial, of the path, could hardly 
 be explamed on any other supposition". 
 
 The method by which Hannibal repaired the path, Ius\ 
 ' buildmg up the ])rccipicc,' may be cx],lained without dif- 
 ficulty; for the words describe gr.aphically the means now 
 resorted to m the Alps to construct or re,,air a path in 
 Bimi ar places. The prccii.icc which had fallen away would 
 not have been one of solid rock, but a very steej, declivity 
 of earth, or, what is perhaps more likely, one of earth and 
 loose rocks mingled together. These precipices are neces- 
 sarily, from the nature of their formation, loosely coherent 
 and very subject to disintegration : the water, in times of 
 rain, easily penetrates them ; the cohesion of their parts is 
 thus destroyed or weakened ; and, owing to the steepness 
 of the declivity, the outer surface of the precipice breaks 
 up, and a landslip is the result 2. Such precipices as these 
 always deviate considerably from the vertical : a sheer pre- 
 cipice must be composed of solid rock. Any exception to 
 
 1 Livy is q.ilto plain l.cro: Iio 6!>js th.it the c.iuso of the accident was 
 tt 'lapsus terra,' or landslip. 
 
 2 A part of the road over the Stclvio jia-^s is mentioned as sufTerinc in 
 
 this manner. ' Tlio steep sides of the mountains along which the road is 
 
 earned, are of a light crumbling soil, in which arc en.bcdded rocks and 
 
 stones. Heavy rains produce great injury to this part of tho road, bv 
 
 washnig away tho soil and bringing down the rocks. Near Trafoi, tho 
 
 road, for a length of about two English miles, w.is nearly destroyed bv 
 
 each accidents, and a new ono on the opposite bank of the stream 
 
 was m consequence opened in 184G.'— .Murray's TTandVook for SoulUrn 
 Uermani/, '' 
 
 fe-i 
 
 Ti 
 
 m 
 
 ^ir 
 
 
 < I 
 
 - V -11 
 
 -I 
 
56 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 lliis rule is very rare iiulccd. Consequeiilly, wlion Han: 
 nibal set his men to work to construct a new path, footing 
 for the workmen to stand might be found witliout much 
 diffieu ty upon the face of the precipice. In their case also 
 It would be more easily penetrable by the feet, in conse- 
 quence of its having been recently broken up; so that 
 none of the surfiiee near the line of the destroyed path 
 could liave been hard, but the men might have buried 
 their feet in the loose ground, and tlms secured a tolerably 
 firm position. ' 
 
 x'i'iio way in which the precipice was 'built up,' and the 
 patli made or repaired, would be such as is now employed 
 I\Ieu would be stationed on the declivity, a fc^Y yards 
 below the line of the broken path ; and from thence, with 
 the aid of materials readily found, such as small rocks 
 stones, and earth, they would build up a kind <.f terrace to 
 the level of the old path, and so restore to it the width it 
 had lost'. As the path is only said to have been made 
 impracticable by the landslip for elephants and beasts of 
 burden, but not for men, we may suppose that some 
 assistance might be rendered from it to those working 
 
 The annexed figures, representing profiles or sections of tl.o dccli. 
 T.ty, may assist in rendering tho mode of operation clearer. Fi- 1 rc- 
 prescnta a place ^vhere the path was partially carried a^ay, AIJ being tho 
 w.dth of the remnant of the path. In fig. 2, the damage is seen remedied 
 by means of a terrace (represented by tho dark part UCV), AJIC being 
 
 boir. n "'" "'T'"'^ ■'■■'"■• ^" '■'- 3. ""> I-->"' iB 8uppose.l to have 
 been .,l,olIy carr.ed away, tho point A marking its former position. 
 In fig. 4, a nc^y path, whoso width is AC, is seen forn.cd by means of a 
 
 Fis. 1. 
 
 / 
 
 K. 
 
 A 
 
 HANNIBAL'S PASSAGE OF Tllli: ALPS. 
 
 57 
 
 below, that materials mi^lit have been furnislicd to tlicni 
 from above, and some help given to tlieni in their arrange- 
 ment. Such a work must, nevertheless, have been long and 
 tedious : and when it is considered, that these terraces had 
 to be raised along the mountain-side for a length of nearly 
 three hundred yards at leasts (for the i)ath was pro])ably 
 either destroyed, or more or less damaged, for the whole 
 of the length mentioned by Polybius,) and also to l)e made 
 sufficiently wide and strong to support the ponderous bulk 
 of the elephants, it will readily be understood how the 
 greater part of three days Mas occupied in their con- 
 struction. 
 
 No vestiges of Ilannibars work can now be expected 
 to remain. It is only the general features of the place 
 which could have continued unaltered to the present day. 
 AVhat its nature was has been now suflieiently explained. 
 Its position also with respect to tlie suinniit of the pass 
 may be easily deduced within certain limits. For Han- 
 nibal arrived at the broken patli before tlie first half of 
 the eleventh day had claj)sed ; and it was on the same day 
 that he began his descent. A condition of the following 
 nature will consequently result^: 
 
 * At a place, within half a day's march of the summit 
 of the pass, on the Italian side, the path must lie along a 
 precipitous mountain-side liable to be broken up for a 
 length of nearly 300 yards ; and the country in the vicinity 
 should be of a nature adapted to Polybius' account of the 
 events which took place in the neighbourhood of the broken 
 path.* 
 
 What these occurrences were will now be considered. 
 When Hannibal first perceived that his path was destroyed, 
 and that it was no longer possiljle for him to descend by 
 the regular way, he attemi)tcd to proceed by taking a 
 circuit, and so avoiding the broken ground. An attempt 
 of this nature, among precipices, seems to have bordered 
 upon temerity. Yet Hannibal probably did not intend to 
 
 1 If tlio path was in zigzag, its length would havo exceeded tho 300 
 yards, tho extent of tho hroken ground. 
 ^ Condition IX. 
 
 i 
 
 
 .'■* 
 
 i» : 
 
 f >. 
 
..4 
 
 
 
 58 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 strike out a ncAv line of route for any Icngtli of distance; 
 but merely to deviate so far from the ordinary way, as 
 was sufKcient to enable him to go round the broken 
 ground, and then to fall again into the regular path. Be- 
 sides, unless he adopted this expedient, he nuist have been 
 obliged to remain (as he ultimately was obliged) for a long 
 time upon the snowy ground, until the broken path coidd 
 be reconstructed. This alternative was in itself almost a 
 desperate one : for the condition of the army nuist have 
 been now seriously impaired, and their provisions nearly 
 gone, while their courage had again left them when they 
 saw their course arrested. 
 
 Hannibal preferred the alternative of inuuediate peril 
 
 to that of lengthened hardship. His attempt to effect a 
 
 circuit was, however, frustrated. The course he had se- 
 
 lected brought him upon a very steep declivity of snow, 
 
 where a mass that was hard and compact, and which l^oly- 
 
 bius says had remained there since the previous winter, was 
 
 thinly covered with snow that had freshly fallen. The 
 
 existence of this mass of old snow gives some clue as to 
 
 one probable feature of the ground where the circuit was 
 
 attempted. It seems clearly an isolated patcli of snow 
 
 that Polybius speaks of. We have no reason for ima- 
 
 gmuig that the line of perpetual snow extended, even in 
 
 those times, when the climate of the Alps was severer than 
 
 It IS at present, as far down as this place would probably 
 
 be. Fresh snow indeed then covered the ground from 
 
 the summit of the pass to this neighbourliood, and con- 
 
 cealed perhaps in the higlier parts of the pass several 
 
 patches of old snow like the one of which Polybius speaks, 
 
 but which, as the fresh snow lay more thickly upon theni 
 
 in consequence of the greater elevation, may have escaped 
 
 notice. * 
 
 Now the places where patches of i)ermancnt snow lie 
 below the line of perpetual snow, (above which, all the 
 mountains, sheer precipices excepted, are buried beneath 
 It) are principally gullies in the mountain-sides, into the 
 hollows of which the snow naturally drifts, and which occa- 
 sionally form tlic channels of avalanches. In consequence 
 ot the fall of one of these, old snow may sometimes be 
 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 69 
 
 found, even at the end of summer, at a very low levcP. 
 This circumstance gives us some reason to suspect that it 
 was by a gully that Hannibal attempted to make his 
 circuit ; a suspicion strengthened by the fact that such a 
 course would be the most natural one to adoi)t. AVhen- 
 ever a man, descending a mountain where there is no 
 track, finds his progress arrested by arriving at the edge 
 of a range of precipices, he naturally keeps along their 
 summit, till he finds a gully breaking through them from 
 above, and affording him a tolerably safe passage down to 
 their feet. A course of this nature would not improbably 
 be that adopted by Hannibal: although it must, in his 
 case, have been of an easier kind than is usual, as he would 
 never have attempted to bring his whole army, men, 
 animals, and baggage, down by such a line as might be 
 taken by an unencumbered traveller. 
 
 AVhen HannibarH attemj)t to make a circuit was frus-^ 
 trated, and he found himself obliged to halt, and repair 
 the broken i)ath, he is said to have encamped, Trepl Ttju 
 pdx^v^ having cleared away the snow upon it. This im- 
 plies that he not only encami)ed near it, but also upon it ; 
 and the word ircpi lias therefore been translated * about.' 
 Hannibal encamped on and around this fyaxis* Now />a;(t9 
 signifies the back or crest of a ridge, * dorsum mentis.* 
 Yet Polybius has previously mentioned no /ja'x'? whatever. 
 AVe first read of it, in connexion with the encampment, as 
 TTjv pdxiv; neither has anything gone before, that neces- 
 sarily implies such a feature in the country ; for the preci- 
 pice, along which the path lay, could not well be described 
 by the word /^«x*^' "^^ ^^^^ ^^ possible to encamp upon it. 
 This ridge-back had possibly, owing to Polybius' being 
 personally ac(iuainted with the country, been present in 
 his mind from the time he first brought Hannibal down to 
 the broken path ; and this may have led him inadvertently 
 
 ^ The lowest situation in which I ever ol)Scrvc{l old snow, was when 
 crossing, in the middle of August, 1850, a gully in the gorge of the Inn, 
 between ^Lirtinsbruck in the Engadinc, and Finstermiinz in the Tyrol. 
 It lay at an elevation of less than 3.'00 feet above the sea, and about COOO 
 below the lino of j)orj)etual riiow. 
 
 
 r ; 
 
 i 
 
 
 i^'" \ 
 
 u 
 li 
 
GO 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S rASSAGE OF THE ALPS. 
 
 61 
 
 to make use of the article in connexion Avitli the /^avt?, 
 -svlien the subsequent course of his narrative required its 
 \ mention. However this may be, the account naturally 
 leads us to look, immediately above the broken path, for an 
 encampment for the Carthaginians, partly extending over 
 a ridge-back. 
 
 ^ As soon as the path was in a sufTicicntly advanced stage 
 of completion to allow the horses and baggage-animals to 
 pass, they were immediately brought down, accompanied, 
 no doubt, by a large portion of the army. Anotlier en- 
 campment was then made below the place where the path 
 had been destroyed, and tiie animals were let loose to find 
 pasturage, this lower district being quite free from snow. 
 This account suggests to us the presence here of a large 
 tract of open pasturage. How far it lay from the broken 
 path is not stated, but it probably was at no great distance 
 from it. This open tract of country is the last charac- 
 teristic of the neighbourhood of the place where the path 
 was broken up, which may be conjectured, from the narra- 
 \ tive of Polybius, to have existed. What is intimated, in 
 addition, by the account of tlie attenq)t at a circuit audits 
 frustration, and of the encampment about the pa\i^, has 
 been already considered above. 
 
 Only one condition now remains to be deduced. It is 
 supplied by the fact, that tlie part of the plains into which 
 Hannibal entered, after his passage of the Alps had been 
 cfFected, was occupied by the Taurini. There is, however, 
 one sentence in Polybius, which seems, at first sight, to 
 imply something different. In Chap. 56 it is stated, that, 
 after Hannibal had passed the Alps, ' he descended boldly 
 into the plains of the Po, and the territory of the Insu- 
 brians.' From this it might be imagined that he emerged 
 at once from the Alps into the country of the Insubrians, 
 the modern Milanese. This suj)position would make him 
 pass the Alps as far eastward as the St Gotliard, or at least 
 as the Simplon, both wliich suppositions are utterly ini- 
 l)robable^ But there is little doubt, that tlie passage 
 
 1 According to Ptolemy, Novara was a town of the Insubrians. Tbo 
 river Sesia may thus liavc been tlicir western boundary. To the west of 
 
 Ik 
 
 V-i. 
 
 
 
 LI 
 
 •*) 
 
 \ V 
 
 t^ 
 
 ' . w^ * ' ' ' V> ^wm>^r^ ' >! ' « « ■ ' > iiiii . ! ti ^ia^i| r i».i> 
 
 above cited, concerning the Insubrians, is one of Polybius' 
 succinct accounts, or sunuuaries before mentioned, and 
 that it merely states the direction and end of a march, the 
 details of which arc afterwards to be given ^ The nature 
 of Hannibal's dealings with the Taurini, which Polybius 
 afterwards relates, perfectly exi)lains what is meant by the 
 word * boldly,' as aj)i)licd to his march into the country of the 
 Insubrians. Though no doubt still anxious not to ])rovokc 
 hostilities, he yet did not avoid them so sedulously as he 
 had done during the j)assage of the Alps, where he only 
 seems to have fought when driven to extremity. From the 
 foot of the Alps a new and bolder system of action was at 
 once adopted. 
 
 It is not, however, necessary to take for granted that 
 the passage relating to the Insubrians is a mere summary. 
 It will appear without difficulty, from other considerations, 
 that the country of the 'J'aurini was the first part of the 
 plains which Hannibal entered after he had crossed the 
 Alps. For, from the passage where the name of the 
 Insubrians occurs, we may conclude, that, wherever it was 
 that Hannibal emerged from the Alps, he descended from 
 that point into the plains, Karyfje roX/jLijpw^ ei? rd ire pi rov 
 UdSov Tre^ia, That he should ever afterwards be close to 
 the Aljjs is not likely. Yet, in the commencement of the 
 60th chapter, (the first part of the history relating to Ilan- 
 nibal after the digression beginning in the middle of the 
 56th chapter) he is spoken of as being encamped vtt avTt]v 
 rriv TTapwpciav tcou ^^AXirewvy under the very slope of the Alps, 
 at the very roots of the mountains Avhen they first spring 
 from the plains. From this mention of the Alps we are 
 led to conclude, that the encampment here spoken of was 
 made be/ore he descended into the plains ; that it should, 
 in fact, have been made when he first came out from the 
 Alps, and previously to his entry into the Insubrian country. 
 This becomes certain from a further examination of the 
 
 tbo Sesia lay Ibc Libui or Libicii, wbo arc distinguisbed by I'olybius (Lib. 
 II. c. 17), from tbo Insubrians. 
 
 1 In order tbat tbo detailed narrative of a marcb may bo clear, it is 
 plainly necessary tbat its direction and object ebould bo previously men- 
 tioned. 
 
 I -i^ 
 
 s*. 
 
 
62 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 63 
 
 60th chapter ; for there the encampment just mentioned is 
 said to have been made, ixerd tyju elajioX^v, 'after the 
 entry ' (the entry into Italy, that is to say, as appears from 
 the els 'iraXiav just preceding). Yet this cannot refer to 
 some indefinite date after the entry into Italy, but must 
 mean the actual time when he first entered the country, 
 the period when he reached the commencement of the 
 plains. The account of the losses which the army had 
 suffered, and the hardships that they had endured during 
 the passage of the Alps, — this account, immediately fol- 
 lowing and connected with the statement of the encamp- 
 ment, also tends to fix the place of that encampment at the 
 point of emergence from the Alps. Besides, as the army 
 remained encamped for some time for the express purpose 
 of recovering the strength it had lost in consequence of the 
 hardships it had endured in the Alps, it seems to follow 
 that this encampment should have been made as soon as 
 they <jame out from the mountains. 
 
 Mera ^e rayra, ' after these things,' as soon, that is to 
 say, as the strength of the army was restored, and they 
 were about to leave their encampment, and advance, their 
 transactions with the Taurini commence. Now the Tau- 
 rini are introduced as dwelling near the slope of the 
 mountains, or at their foot, oi Tvy')(avovai Trpos Trj irapiopeia. 
 KaroiKovvre^. What then was this particular irapwpeia ? It 
 was clearly the Trapwpeia which is mentioned at the begin- 
 ning of the chapter, and beneath which Hannibal encamped 
 when he first issued from the mountains. That the ex- 
 pression, oi Tvyx^dvovcri, k,t.\, should merely mean that 
 they dwelt somewhere at the foot of the Alps, is out of 
 the question. Besides incurring the objection of irrele- 
 vancy, such an interpretation would not even define the 
 position of the Taurini ; since they might, as far as that 
 description of their country is concerned, have lived any- 
 where along the base of the Alps, from the southern fron- 
 tier of Piedmont to the eastern extremity of Friuli. The 
 phrase also, 'they happened to dwell at the foot of the 
 mountains,' is clearly employed for the purpose of ex- 
 plaining how they lay upon the line of HannibaPs intended 
 march, which was directed, as Polybius had before inti- 
 
 mated, towards the country of the Insubrians. They hap- 
 pened to lie in his way : he could not avoid passing through 
 their country, and having some transactions with them. 
 
 The Trapoipeia, therefore, by which the position of the 
 Taurini is defined, being necessarily identical with the 
 Trapwpeta beneath which Hannibal encamped; and this 
 latter irapulpeia being situated at the point where Hannibal 
 first emerged from the mountains; it may be concluded, 
 that the Taurini occupied the plains into which Hannibal 
 first entered after his passage of the Alps. This conclusion 
 is also, as has been seen, strengthened by the fact of 
 Polybius' employing the word Ti/7^aVoi;cn, as if expressly 
 to intimate that the Taurini lay in the way of Hannibal as 
 he marched towards the country of the Insubrians ^ 
 
 Indeed, had Hannibal first descended into the Tnsu- 
 brian territory, the attack on the Taurini would be very 
 difficult to explain. That the Taurini were at war with the 
 Insubrians, the allies of Hannibal, is no sufficient reason. 
 This circumstance is mentioned by Polybius to account 
 for the Taurini refusing the overtures of Hannibal, but 
 would not be a sufficient cause to induce the Carthaginian 
 general, when he had once arrived in the Milanese, and 
 was about to march upon Southern Italy, to go so far out 
 of his way as Turin, for the purpose of attacking, with an 
 army ill adapted to bear further diminutions, a people with 
 whom there was no necessity for hostilities. It is, how- 
 ever, unnecessary to dwell upon the* unreasonableness of 
 such a supposition. That Hannibal descended at once from 
 the Alps into the country of the Taurini is sufficiently plain, 
 without the aid of any such considerations : and the impro- 
 bability of a contrary supposition need not be canvassed 
 at any length. 
 
 The following, then, is the last condition for the deter- 
 mination of Hannibal's route over the Alps^: 
 
 * The plains into which the road over the pass enters, 
 when it emerges from the mountains on the side of Italy, 
 must anciently have been inhabited by the Taurini.' 
 
 1 A country which we know he reached after leaving Turin, and 
 probably never reached but once. 
 
 2 Condition X. 
 
 1 
 
64 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 65 
 
 The examination of Polybius' narrative of Hannibal's 
 march, from the confluence of the Bhone and Isere to the 
 commencement of the plains of Italy, being now concluded, 
 all that remains to be done is, to present in one view the 
 principal points which have been elicited from it : to give, 
 in the first place, the journal of the march ; and then to 
 collect together the various conditions which have been 
 deduced, for the purpose of assisting in the identification 
 of the pass, by which Hannibal crossed the Alps. 
 
 Starting, then, from the confluence of the Rhone and 
 the Isere (on the left bank of the latter river) the journal 
 of the march will be as follows : 
 
 I. The ten days* march to the town of the Allobroges, 
 including the halt in that place. 
 
 1—7. Hannibal marches up the left bank of the Isere to the 
 neighbourhood of the defile leading to the town : he finds 
 that the heights commanding the defile are held by the 
 Allobroges, and sends forward a party of Gauls to learn 
 the plans of the enemy. 
 
 8. The Gauls having returned with the requisite intelligence, 
 
 Hannibal advances to the entrance of the defile and there 
 encamps : he seizes in the night the abandoned i^osts 
 of the Allobroges. 
 
 9. The defile is passed with much loss, and the town captured. 
 10. The army remains in the neighbourhood of the town. 
 
 IL The fifteen days' march across the Alps, from the 
 town of the Allobroges to the commencement of the plains 
 of Italy. 
 
 1. Hannibal begins his march from the town : he meets, either 
 on this evening, or on the following morning, a deputation 
 of the Alpine Gauls : he makes a treaty with them, and 
 subsequently employs them as guides. 
 2, 3. Hannibal proceeds on his march under the guidance of the 
 Alpine Gauls. 
 4. The Carthaginians are attacked by the Alpine Gauls, in 
 front and rear, while passing through a defile : the attack 
 on the rear is repulsed, but great loss is suffered in the van. 
 Hannibal eventually gains possession of a commanding 
 position on the heights, near a certain 'strong white 
 rock/ He remains during the whole night upon these 
 heights with half his army, apart from his cavalry and 
 
 . > 
 
 baggage-animals, whose passage through the defile he thus 
 
 protects. 
 5. Hannibal rejoins his cavalry and baggage, and continues 
 
 his march towards the summit of the Alps. 
 6 — 8. The army proceeds on its march. It is harassed at in- 
 tervals, during the march from the Rock, by predatory 
 
 attacks on the part of the barbarians. 
 9. On this morning the summit of the pass is gained, and the 
 
 army encamps for the sake of rest, and in order to allow 
 
 stragglers to rejoin it. 
 
 10. The army remains encamped on the summit of the pass : 
 
 the troops become greatly discouraged. In order to remove 
 .their despondency, Hannibal brings them to a point within 
 sight of the plains of the Po, and indicates to them the 
 quarter in which Rome itself lies. 
 
 11. The army begins to descend. The descent is precipitous, 
 
 and the path is concealed by the snow. Multitudes, both 
 of men and animals, miss their footing, and are lost down 
 the precipices. A place is ultimately reached, where the 
 path is destroyed, and the march arrested. Hannibal 
 attempts to continue his progress by making a circuit. 
 The attempt is frustrated in consequence of the snow. 
 Hannibal encamps about a ridge-back above the broken 
 path, which he begins to repair. 
 
 12. The construction of the new path continues. On this 
 
 morning, or on the previous evening, it is made practicable 
 for horses and baggage-animals : they are brought down 
 and left to seek forage in the pasture-grounds which lie 
 below the broken path and are free from snow. 
 
 13. The path is made practicable for the elephants. They 
 
 are brought down, and the whole army is again collected 
 together. 
 
 14. The army marches towards the plains of Italy, exposed 
 
 only to slight and furtive attacks. 
 
 15. The army continues its march, arrives at the commence- 
 
 ment of the plains of Italy, and encamps at the roots of 
 the Alps. 
 
 The conditions for the identification of the pass are 
 these : 
 
 I. The commencement of the ascent of the Alps must be 
 
 situated at a distance of about 100 Roman miles from 
 the junction of the Rhone and the Is^re, reckoned along 
 the left bank of the latter river. 
 
 II. The length of the route over the Alps, beginning at the 
 
 5 
 
66 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS, 
 
 67 
 
 commencement of the ascent of the mountains, and ter- 
 minating at the commencement of the plains of Italy, 
 must be about 150 Roman miles. 
 
 III. At the commencement of the ascent of the Alps, a defile 
 
 must be found, the character of which, and of the surround- 
 ing country, is in accordance with the events related by 
 Polybius to have occurred in that place. 
 
 (Minor conditions included in the above.) 
 
 (1) The defile must be commanded by certain heights of con- 
 
 siderable elevation. 
 
 (2) These heights must not be easily accessible from below 
 
 the defile. 
 
 (3) The way through the defile must skirt in some place the 
 
 edge of a precipice. 
 
 (4) Immediately below the defile must be found a place 
 
 where an army as numerous as that of Hannibal could 
 encamp. 
 
 (5) This place of encampment should be visible from the 
 
 heights commanding the defile. 
 
 (6) Above the defile, and near it, must be found an open dis- 
 
 trict, where a town either stands or might have stood, 
 and where an army such as HannibaFs could have 
 encamped. 
 
 IV. At a place nearly half-way, in point of time, between the 
 
 town of the AUobroges and the summit of the pass, a 
 certain * strong white rock' must be found ; and the 
 adjacent country must be in accordance with the events, 
 which are recorded by Polybius to have taken place in 
 the neighbourhood of this rock. 
 
 (Features of the country in the neighbourhood of the 
 Eock.) 
 
 (1) A difficult and precipitous ravine, through which the road 
 
 must pass. 
 
 (2) A range of acclivities overlooking the road, and liable 
 
 to be swept, in one or more places, by rocks set in 
 motion from above. 
 
 (3) A large extent of practicable ground above these acclivi- 
 
 ties. 
 
 (4) A position on this ground, at some distance from the road, 
 
 where 20,000 men could be stationed, so as to protect 
 completely the flank of an army marching along the 
 road beneath, and such as would not expose themselves 
 to be attacked at any serious disadvantage, by an 
 enemy lying beyond them relatively to the road. 
 
 
 V. A place suitable for the encampment of Hannibal's army 
 
 must be found at the summit of the pass. 
 
 VI. From a point, probably not far from the road over the sum- 
 
 mit of the pass, but yet not upon the ground where the 
 Carthaginians would encamp when they first reached the 
 summit, the plains of the Po, and, in all probability, the 
 Apennines also, ought to be visible. 
 
 VII. The commencement of the plains of Italy must be less 
 
 than three days' march from the summit of the pass. 
 
 VIII. The first part of the descent from the summit of the pass 
 
 on the Italian side must be of a precipitous character. 
 
 IX. At a place, within half a day's march from the summit of 
 
 the pass, on the Italian side, the path must lie along a 
 precipitous mountain-side, liable to be broken up for a 
 length . of nearly 300 yards (English), and the country 
 in the vicinity should be of a nature adapted to Polybius' 
 account of the events which took place in the neighbour- 
 hood of the broken path. 
 
 X. The plains, into which the road over the pass enters, when it 
 
 emerges from the mountains on the side of Italy, must 
 anciently have been inhabited by the Taurini. 
 
 ^ 
 
 5 — 2 
 
\ 
 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 69 
 
 CHAPTER VI. 
 
 Improbability of Hannibars having crossed the Alps by the Great or 
 Little St Bernard, or by the Mont Genevre. — Probability of his 
 having crossed the Mont Cenis. — Determination of the point where 
 Hannibal would have left the Isere. — The oldest line of road known 
 over the Mont Cenis traced and described. — The Little Mont Cenis 
 selected as the pass which Hannibal crossed, in preference to the 
 Great Mont Cenis. 
 
 SEVERAL passes have been suggested as having each 
 their claims to be considered as the pass which 
 Hannibal crossed ^ Four only, however, seem to have, 
 even at first sight, any considerable probability in their 
 favour : the Great and the Little St Bernard, the Mont 
 Cenis, and the Mont Genevre. The Great St Bernard 
 might perhaps also be left out of consideration : but as the 
 examination which leads to the rejection of three out of 
 che four passes named will not be long, it has been thought 
 better to include it among the number. 
 
 In the endeavour to identify the pass which Hannibal 
 crossed, the first question to be asked is: What are its 
 distinguishing characteristics? Are there any circum- 
 stances connected with it of an extraordinary nature, such 
 as are rarely to be found on any of the passes of the Alps; 
 and such as, if not found on any particular pass, raise 
 great difficulties, if not impossibilities, against the adoption 
 of that pass ? Now, there are two circumstances of this 
 kind, which have been deduced from Poly bins' narrative, 
 and are embodied in Conditions VI. and VII. That Italy 
 should have been visible from the summit of the pass, and 
 that the commencement of the plains of the Po should lie 
 
 1 The names of these passes, and of the advocates for each, may be 
 found in the Appendix on HannibaFs passage in Ukert's Oeographie der 
 Griechen und Homer, 
 
 within less than three days' march from the summit, — these 
 are the prominent and distinguishing characteristics of the 
 pass which Hannibal crossed. 
 
 But, of the four passes named, it is only from one, the 
 Mont Cenis, that Italy can be discerned. With respect to 
 the other three, neither from the roads themselves, nor 
 from any neighbouring eminence, would it be possible to 
 see any part of Italy, as distinguished from the Alpine 
 country. Nothing but ranges of mountains, and perhaps 
 some valley locked in on all sides by them, would meet the 
 eye. From the Mont Cenis alone could Hannibal have 
 pointed out to his soldiers the plains of the Po^ 
 
 The great proximity also to the plains of Italy of the 
 summit of the pass which Hannibal crossed, its second 
 distinguishing characteristic, is only to be remarked on 
 the pass of the Mont Cenis. The distance between the 
 summit of the pass which Hannibal crossed and the com- 
 mencement of the plains was less than three days' march. 
 Now from the point where Hannibal left the banks of the 
 river (the Isere) to the commencement of the plains, the 
 distance is given as 150 Roman miles. From the town of 
 the AUobroges, therefore, from which the fifteen days' 
 march across the Alps is reckoned, the distance to the 
 commencement of the plains would be rather less ; perhaps 
 140 Roman miles or a little more. Of the fifteen days 
 which Hannibal employed in the passage of the Alps, only 
 eleven were occupied in marching : nearly two were spent 
 on the summit of the pass, and fully two lost, during the 
 descent, at the broken path. Consequently, as Hannibal 
 marched 140 Roman miles, or perhaps a little more, in 
 eleven days, his rate of marching in the Alps would be 
 about thirteen Roman miles a day. On his descent he 
 might march rather more rapidly than on his ascent : but 
 the difference would not be considerable, for the vigour of 
 the army was then much diminished. Three days' march, 
 
 1 Der Cenis ist auch der einzige Berg, uber welchen eine grosse 
 Strasso fuhrt, Ton dem man an mehren Stellen die Ebene Oberitalien's 
 sehen kann, was weder vom grossen und kleinen Bernhard, noch vom 
 Simplon moglich ist.— Ukert, Geographie, n. 2, p. 600. (The Genevro 
 Bhould also have been mentioned by Ukert among these latter passes.) 
 
/ 
 
 70 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 71 
 
 at the rate of thirteen Roman miles a day, will be sufficient 
 to allow for the distance between the summit of the pass 
 and the commencement of the plains, especially as not 
 quite three days were spent on the actual march. It may 
 therefore be inferred, that the distance between the 
 summit of the pass and the commencement of the plains 
 of Italy did not exceed forty Eoman miles. 
 
 The roads from the Great and the Little St Bernard 
 into Italy unite at Aosta, and emerge into the plains of 
 Italy near Ivrca, which may for these two passes be fixed 
 upon as the commencement of the plains ^ The roads 
 over the Mont Cenis and the Mont Genevre unite in like 
 manner at Susa, and enter the plains nearly at Avigliana*; 
 although Eivoli is sometimes fixed upon as the place where 
 the plains may rather be said to begin. Yet the irregu- 
 larities in the ground below Avigliana cannot be considered 
 as parts of the mountains: the conspicuous Monte Piri- 
 chiano, crowned by the monastery of San Michele, and at 
 the foot of which Avigliana lies, stands out plainly when 
 seen from any point in the neighbourhood of Turin, as the 
 last of the Alps. There is, besides, ample room about 
 Avigliana for an encampment; and if Hannibal had ad- 
 vanced further with his enfeebled army, he would have 
 lost the protection of the mountains behind him, and ex- 
 posed himself to attack from the Taurini before he was 
 well prepared for hostilities. 
 
 Ivrea, anciently Eporedia, was a Eoman station, and 
 lay, according to the Itineraries, at a distance of 46 miles 
 from Aosta. From Aosta to the summit of the Great St 
 Bernard (Summus Penninus) was a distance of another 
 15 miles 2: so that the whole distance from Ivrea to the 
 summit of the Great St Bernard was 61 Eoman miles. 
 
 1 See Penny Cyc. Art. * Ivrea.* Blaev, Theatrum Statuum Regice Cel- 
 situdinis Sabaudice Duels, Art. * Eporedia.* 
 
 2 See Murray's Hand-Book for Northern Italy, ' Avigliana.' 
 
 3 According to the Roman Itineraries, 25. The Government Itinerary 
 of the Sardinian States (1839) gives, however, 10 Piedmontese or 16^ 
 Roman miles, as the distance from Aosta to the Hospice of the Great St 
 Bernard, so that the xxv miles of the Roman Itinerary is probably a mis- 
 take for XV. 
 
 Mi 
 
 
 ' I 
 
 < 
 
 V 
 
 From Aosta to Arebrigium (generally identified with 
 Pre St Didier, but perhaps rather Arpi on the road from 
 Morgex leading to La Thuile and the Little St Bernard 
 by the 'Camp of Prince Thomas'), the Itineraries give 
 25 miles. From Arebrigium to Artolica (La Thuile) is a 
 distance of another 6 miles: and from La Thuile to the 
 summit of the Little St Bernard (Alpis Graia) there are 
 about 5 miles more. From Ivrea to the summit of the 
 Little St Bernard is consequently, according to the Itine- 
 raries, a distance of 82 Eoman miles. 
 
 Avigliana is the * Ad Fines ' of the Itineraries, and lay 
 at a distance of 24 Eoman miles below Susa. From Susa, 
 taking the old road over the Mont Cenis, the distance to 
 La Novalese is about 5 English miles : from La Novalese 
 to La Grande Croix, at the southern extremity of the 
 plateau of the Mont Cenis, the distance is about 6 English 
 miles. To the centre of the lake of the Mont Cenis (which 
 lies nearly in the middle of the plateau of the Great Mont 
 Cenis, and from which the plateau of the Little Mont Cenis 
 branches off at right angles to the other plateau ^ the 
 distance will be nearly 3 English miles more : so that, from 
 Susa to the central point of the plateaux of the Mont 
 Cenis, the distance is nearly 14 English, or about 15 
 Eoman miles. From Avigliana, then, to the central point 
 of the summit of the Mont Cenis, the distance is 39 
 
 Eoman miles. 
 
 From Susa to Ad Martis, the modern Oulx, the Itine- 
 raries give 16 Eoman miles, and 9 more to Gesdao or 
 Cesanne. From Cesanne to the central point of the plain 
 of the Mont Genevre the distance may be taken at 4 
 Eoman miles. From Avigliana to the summit of the Mont 
 Genevre the distance is thus 53 Eoman miles. 
 
 In all these cases it appears, by comparison with the 
 
 1 There are two ways of ascending the Mont Cenis from the side of 
 Savoy: (1) by the road over the great Mont Cenis, the present high 
 road, which leaves the valley of the Arc at Lanslebourg ; and (2) by the 
 road of the Little Mont Cenis, a mule-path, which turns off from the valley 
 of the Arc at Bramans. Both these passes are of nearly equal height. The 
 two roads unite on the summit of the mountain, and the descent into Italy 
 is the same for both. 
 
 A 
 
V 
 
 72 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 present distances, that there is no material error in the 
 Itineraries, excepting in the instance already noticed. 
 
 From the summits of the respective passes to the com- 
 mencement of the plains the distances therefore are : 
 
 For the Great St Bernard 61 Roman miles. 
 
 For the Little St Bernard 82 „ „ 
 
 For the Mont Cenis 39 „ „ 
 
 For the Mont Genbvre 53 „ „ 
 
 And the actual distance on the pass which! ^^ 
 
 Hannibal crossed probably did not exceed/ " " 
 
 The Mont Cenis seems here to be plainly indicated as 
 the pass which Hannibal crossed. With respect to either 
 of the St Bernards, the discrepancy in point of distance 
 raises not merely a difficulty, but it may be said an im- 
 possibility to overcome. The fact also, already noticed, 
 of the invisibility of the plains from their summits, opposes 
 another obstacle, equally strong, to the adoption of either 
 of these passes. 
 
 A third objection arises from the circumstance, that it 
 was not the Taurini, but the Libui, who would have in- 
 habited the plains into which Hannibal entered on his 
 emergence from the Alps, had he crossed either the Great 
 or the Little St Bernard. Yet the Libui, although a tribe 
 known to Polybius, are never mentioned by him in his 
 account of Hannibal's passage of the Alps, the narrative 
 necessarily implying that the Carthaginians found the Tau- 
 rini in their way as soon as they issued out from among 
 the mountains, and began their march to the Insubrian 
 territory. 
 
 These three objections seem fatal to the Great and the 
 Little St Bernard. The argument from discrepancy of 
 distance presses also strongly, though perhaps not conclu- 
 sively, against the Mont Genevre, from which the plains 
 of Italy are, besides, invisible. The route of the Mont 
 Genevre is, in addition, at variance with Condition I ; for 
 Hannibal would, on the supposition of his having crossed 
 the Genevre, have to leave the Isere near Grenoble, and 
 penetrate into the mountain-valley of the Drac. But the 
 distance between Grenoble and Valence, which is of the 
 same length as the distance between Grenoble and the 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 73 
 
 confluence of the Isere and the Rhone, is only 96 kilo- 
 metres, or about 66 Roman miles; a distance too widely 
 removed from the 800 stadia or 100 Roman miles of Poly- 
 bius, to be admitted. 
 
 Of the four passes mentioned, three seem thus to be 
 excluded from consideration by strong improbabilities. 
 The Mont Cenis remains alone with likelihood in its favour. 
 The next thing, therefore, to examine is, whether the 
 characteristics of this latter pass are completely in accord- 
 ance with the conditions derived from the narrative of 
 Polybius. Should they prove so, the adoption of the Mont 
 Cenis as the pass which Hannibal crossed, seems an inevi- 
 table conclusion ; for the negative evidence in its favour, 
 arising from the great improbabilities with respect to the 
 other passes, is exceedingly strong. We shall consequently 
 jiow proceed to apply systematically to the Mont Cenis 
 all the conditions which have been deduced from the dis- 
 <jussion of Polybius* account of Hannibal's passage of the 
 
 Alps. 
 
 For this purpose it will first be necessary to trace the 
 oldest line of road over the Mont Cenis, which rarely ever, 
 when the mountains are once entered, coincides with the 
 modern road. The marshes and defiles, which the modern 
 road. has been enabled to traverse, by the aid of dykes to 
 restrain the rivers, or gunpowder to blast away the rocks, 
 were, in remote times, obstacles not to be directly over- 
 come. They forced the road away from the river, which 
 .flowed through the valley leading to the summit of the 
 pass, and compelled it to skirt or to ascend the flanks of 
 the mountains at the side of the valley, in search of prac- 
 ticable ground on which it might run. 
 
 From the confluence of the Rhone and the Isere as far 
 as Grenoble, there can be no doubt about the direction 
 taken by a road making for the Mont Cenis by the left 
 bank of the Isere. It must always have followed the river, 
 nearly in the line of the modern road, leaving it probably 
 a little to the left on approaching Sassenage and Gre- 
 noble, where there is some marshy ground which might 
 cause it to recede from the Isere. From Grenoble the 
 easiest way of proceeding to the Mont Cenis would still be 
 
74 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 \. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 75 
 
 « 
 
 to pursue the broad vale of Graisivaudan, (as the French 
 part of the Val Isere is called) and so gain the entrance of 
 the valley of the Arc, the river which waters the Mau- 
 rienne and leads to the Mont Cenis. A much shorter, 
 though, at the same time, a much more difl&cult way- 
 would be, to cross the mountains, if possible, between 
 Grenoble and St Jean de Maurienne, and thus avoid the 
 great circuit which the easier road makes. There was a 
 Eoman road which took this shorter course. Remains of 
 it have been found in the neighbourhood of the Col de la 
 Coche, upon the mountains between the Graisivaudan and 
 the valley of the die. It would, therefore, in all proba- 
 bility, strike off from the road along the left bank of the 
 Isere, about twelve miles above Grenoble, and a short 
 distance below Froges, — at the point where the stream 
 descending from the Col de la Coche enters the Graisi- 
 vaudan. 
 
 Having crossed the Col de la Coche, and entered the 
 upper valley of the OUe, its subsequent course, if it was 
 intended to lead into the Maurienne (and this was evi- 
 dently, by its crossing the Col de la Coche, the ultimate 
 design), would be most likely by way of the Col du Glandon. 
 From the summit of this pass it would descend into the 
 valley of the Arvan, and join the other line of road, leading 
 to the Mont Cenis, at St Jean de Maurienne. It is said, 
 and in all probability with justice, as will be seen in 
 a subsequent chapter, that this road formed part of the 
 route followed by Julius Csesar, when he crossed the Alps 
 in haste to intercept the Helvetii, who had broken into 
 Transalpine GauP. 
 
 ^ With this pass from the Graisivaudan to St Jean de Maurienne 
 I am personally unacquainted. It is mentioned in the following note from 
 Pilot's Histoire de Grenoble, p. 7. Grenoble, 1829. 
 
 II y avait une autre route qui traversait les montagnes de Theys pour 
 aller dans la Maurienne. On en trouve encore une partie considerable qui 
 se dirige vers le Col de la Coche ; elle est pavee en blocs bruts de granit, 
 et a pres de dix pieds de largeur. On raconte communement que Cesar 
 est entre dans les Gaules par ce chemin, lorsqu'il partit d'Aquilee. Ab 
 Ocelo, quod est citerioris provincioe extremum, in fines Vocontiorum ulterioris 
 provincice pervenit : inde in Allobrogum fi^ies. 
 
 This route would not be that followed by Hannibal. It 
 leaves the banks of the Isere too soon to be in accordance 
 with the measurements of Polybius ; and the length of the 
 Alpine route would also be too short. Besides, we have no 
 record of any mountains having been crossed by Hannibal, 
 with the exception of the actual pass, * the highest summits 
 of the Alps.' An easier, but longer route, seems to have 
 been the one followed by the Carthaginians, which must 
 also have struck off from the Isere at a greater distance 
 above Grenoble. 
 
 From Grenoble to the confluence of the Isere and the 
 Arc, the valley of the Isere is bounded on the east by a range 
 of hills, lying in advance of the Alps behind, and cut through 
 by the lateral streams which descend to the Isere. At first, 
 on leaving Grenoble, all these hills are connected with the 
 mountains behind, but at Le Cheylas, about 20 miles 
 above Grenoble, they become distinct. Passing through 
 the gorge of Le Fay, which here breaks through the chain 
 of hills, and affords a passage for the diligence-road to 
 Allevard, a broad and rich valley is reached, lying between 
 the hills and the Alps, and extending for a length of rather 
 less than 20 miles, from St Pierre d'Allevard to the 
 marshes where the Arc joins the Isere. The upper or 
 French part of this valley is called the valley of Allevard ; 
 the lower or Savoyard part, divided from the upper by the 
 river Bredaz, and a small tract of hilly ground to the north 
 of that torrent, is called the valley of La Eochette. 
 Through this valley, according to M. Albanis de Beaumont, 
 a Eoman road formerly ran : he also speaks of the former 
 importance of La Eochette, and of the fertility of the sur- 
 rounding country ^ 
 
 1 A Textremite de cette vallee est le bourg de la Eochette, autrefois 
 ires considerable, et ayant titre de yille ; il est memo connu, qu'ancienne- 
 ment ce bourg se trouvoit sur une des voies romaines qui conduisoit d'Ai- 
 guebelle a Vienne, alors la capitale de 1' Allobrogie ; j'ai aussi ete a meme 
 de voir plusieurs titres qui prouyent que cette ville etoit dans le douzieme 
 siecle tres considerable, puisqu*Aymon, seigneur de Mont Maieur, y leva h. 
 cette epoque cinq cents lances et deux cent cinquante cavaliers. La situ- 
 ation de la Rochette est tres agreable, et ses environs, qui sent tres fer- 
 tiles, oflfrent une quantite de sites aussi romantiques que pittoresques. — 
 Les Alpes Orecqus et Cottiennes, Vol. II. p. 595. Paris. 1802. 
 
( 
 
 76 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 At Pont Charra, four or five miles above Le Cheylas, the 
 line of hills is again broken through, and the Bredaz issues 
 forth. Through this defile a road runs by way of Mou- 
 taret to Allevard ; but the defile does not afford a con- 
 venient access to La Eochette, the road to which place, 
 from Montmelian, evades the defile by passing over the 
 heights to the north. A road to Grenoble, from La 
 Rochette, would be more likely to take the line of Allevard 
 and Le Cheylas than that of Pont Charra. 
 
 There is also a path from Allevard over the mountains 
 to St Jean de Maurienne. It crosses the Col du Merlet, 
 and descends into a valley which joins that of the Arc at 
 Ste Marie de Cuines. From this tributary valley of the 
 Maurienne a path over a lateral ridge leads to St Jean de 
 Maurienne. This route is probably of some antiquity, as 
 immediately below the crest of the Col du Merlet, on the 
 eastern side, a group of chalets is marked on the Govern- 
 ment map of this country, as the * Granges de la Vieille 
 Route.' It is solely on this account that it is mentioned 
 here ; for it is not likely to have been the route followed 
 by Hannibal. 
 
 The last line of route on the left bank of the Isere, 
 by which the Maurienne could be reached from Grenoble, 
 is the modern road, which follows the river over the 
 frontier into Savoy. At a place called La Chavanne, about 
 half or three quarters of a mile from Montmelian, it falls 
 into the high road from Chambery to Turin by the Mont 
 Cenis, which leads by way of Maltaverne to the entrance 
 of the valley of the Arc. 
 
 Two lines of road have therefore to be considered : 
 
 (1) The road which branches off from the Graisi- 
 vaudan at Le Cheylas ^ and leads by way of Allevard and 
 La Rochette to Aiguebelle. 
 
 (2) The road by Pont Charra, La Chavanne, and 
 Maltaverne, to the same place. 
 
 It is the first of these two roads which is in accordance 
 
 1 The oldest path probably left the Graisivaudan here. The modern 
 diligence-road to Allevard branches off at Goncelin, and has been con- 
 structed, partly by the aid of gunpowder, along the Boutheru side of the 
 gorge of Le Fay. 
 
 \. 
 
 ti 
 
 ii 
 
 {{ 
 
 ^^ 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 77 
 
 with the requisitions of Polybius' narrative. The length of 
 both routes is almost exactly the same ; but the country 
 through which the first route leads is more productive. It 
 is, however, rather the more difficult of the two, though 
 both are partially hilly. Yet the ascent which is requisite 
 on the first road is not very important. The elevation 
 of Allevard, which lies on the highest part of the road, 
 is given as 170 metres, or 560 English feet^ above Gre- 
 noble, which would make it less than 500 feet above Le 
 Cheylas, perhaps not more than 400. 
 
 There appear, however, sufficient reasons for supposing, 
 that in former times, when the country was in its natural 
 state, its barbarous inhabitants, when they wished to go 
 from Grenoble to Aiguebelle, might not only have chosen, 
 but may have been absolutely obliged, to take their way 
 through Allevard and La Rochette. For the road which 
 goes by way of Maltaverne twice crosses great tracts of 
 marsh-land, which in ancient times must have been im- 
 practicable morasses. The first of these lies between Pont 
 Charra and Montmelian, the second between Maltaverne 
 and La Croix d' Aiguebelle. Neither is it possible to skirt 
 these marshes and so avoid them, for they extend, in both 
 cases, to a considerable distance on the right of the road, 
 along the banks of two small streams, which have to be 
 crossed. The first of these streams descends from Malta- 
 verne, and joins the Isere between Pont Charra and Mont- 
 melian ; the second is the Gellon, descending from the 
 valley of La Rochette. In fact, with the exception of the 
 part where the road keeps along the range of hills ex- 
 tending from Maltaverne to La Chavanne and also a little 
 beyond each place, nearly all the route from Pont Charra 
 to La Croix d' Aiguebelle is marshy. There are also small 
 marshes by Maltaverne, on the heights, but they might 
 have been avoided without difficulty. 
 
 The marshes between Montmelian and Pont Charra are 
 thus mentioned in a short notice on the territory of La 
 Rochette 2: 
 
 1 In a small book, called the Guide du Visiteur au 'pays d* Allevard, par 
 P. A. Rigollot Delavaquerie. 
 
 « Richard, Guide en Savoie et en Piemont. 
 
/ 
 
 78 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 79 
 
 * Ketendue de ce territoire est d'environ 20 kilom. ou 
 4 lieues, sur 3 lieues de large. La partie qui longe Tlsere 
 apres le pont de Montmelian est composee d'une terre 
 d'alluvion tres fertile ; il se deploie ensuite sur les bords de 
 risere, en forme de grands marais, une vaste prairie qui 
 descend j usque vers Pont Charas.* 
 
 M. Albanis de Beaumont, in his Description des Alpes 
 Grecques et Cottiennes, also mentions these marshes, as they 
 extend by St Helene du Lac towards Coise and Maltaverne. 
 Speaking of the road between Montmelian and Maltaverne 
 along the crest of the ridge overlooking these marshes, he 
 
 proceeds : 
 
 «Du hameau de la Chavane, (near Montmelian) la 
 route, quoique legerement montueuse, est tres agreable, 
 etant ombragee de chaque cote de gros noyers et divers 
 arbres fruitiers. A une demi-lieue de ce hameau on 
 commence a apercevoir qk et la des bancs de schistes feuil- 
 letes d'une espece argilleuse mais tres tendre ; ces schistes 
 paraissent se prolonger jusqu'a la haute coUine de mont 
 Maieur, situee au sud-est du chemin, dont elle rCest separee 
 que par une suite de belles prairies et de marais qui se pro- 
 longent jusquau lac de S*' Helene; ces prairies ont visiblement 
 ete sous les eaux, et elles ont fait dans tin temps partie de 
 
 ce lac^^ 
 
 All these marsh-lands are now in a great measure 
 reclaimed by means of drainage and embankments : but 
 in ancient times they must have formed, as may be con- 
 cluded from the passages above cited, and from the 
 appearance of the country, a great swamp, which prevented 
 all passage along the left bank of the Isere, between Pont 
 Charra and Montmelian. In order to reach Aiguebelle 
 from the Graisivaudan, it must have been necessary to 
 pass by way of La Rochette ; or else, (which is however a 
 mere supposition, for no road is known to have adopted 
 such a course) to have proceeded from Pont Charra, along 
 the north-western flank of the ridge of Montmayeur. 
 
 But even if Maltaverne were reached, the same diffi- 
 culties would occur again. Another large tract of marsh 
 
 ('I 
 
 defends the entrance of the valley of the Arc. Bourgneuf 
 is already on the fen, and from thence to La Croix d' Ai- 
 guebelle the road lies entirely over marsh-land. A great 
 quantity of ground has indeed been reclaimed here ; yet 
 extensive swamps are still left on the banks of the Isere 
 and the Arc, though both these rivers are now almost 
 entirely confined by dykes ; the vast undertaking of their 
 embankment being nearly completed, down to the place 
 where the Isere enters France \ Fifty years ago, M. Alba- 
 nis de Beaumont remarked the great swamps then existing 
 about the confluence of the Isere and the Arc 2. M. Ber- 
 tolotte also, in the year 1827, took notice of these morasses, 
 * le paludose lande ove F Isera riceve le acque delP Arco ;' 
 and mentions the swampy meadows, * i prati paludosi,' as 
 being found on the banks of the Gellon as far from the 
 Arc as the neighbourhood of Bettonet^. 
 
 It seems clear then that, in ancient times, a large marsh 
 lay between Maltaverne and the entrance of the valley of 
 the Arc ; and it is not likely that any way ever lay across 
 it, before the Romans, in all probability, (as will be seen in 
 a later chapter) constructed a line of road here from the 
 Maurienne to Montmelian. The barbarian inhabitants of 
 the Maurienne, when seeking a way to the Graisivaudan, 
 
 1 Xes AJfZi Grecqites et Cottiennes, p. 691. The author speaks again 
 (p. 690) of the lake of Ste Helene, and the marshes adjoining it. 
 
 1 When these dykes are completed, the high road from Montmelian 
 to Turin will be carried along them by the sides of the Isere and Arc into 
 the Maurienne, and the present road by Maltaverne will be abandoned. 
 The new road was indeed, in the spring of the present year (1853), nearly 
 finished. 
 
 2 A une demi-lieue de Maltaverne Ton traverse le Gelon ; h droite la 
 vue plonge dans la vallee du Beton, et a gauche dans celle de Tlsere, dont 
 on suit les sinuosites de la riviere jusqu'h Conflans, situe a Tentree de la 
 vallee de Tarantaise. Bientot Ton se trouve vis-^-vis du confluent de 
 risere et de TArc, ou le voyageur sentimental voit avec regret les ravages 
 que causent les eaux de ces deux rivieres, particulierement sur la rive 
 gauche de I'lsere, ou il y a des marais d'une etendue de plu^sieurs lieues, dont 
 les exhalaisons ou miasmes qui s'en degagent lors des grandes chaleurs, 
 poussees par les vents nord-ouest, et remontant dans I'etroite vallee de 
 Maurienne, y occasionnent annuellement des fievres intermittentes, et 
 plusieurs autres especes de maladies egalement nuisibles aux habitants. — 
 Zes Alpes Grecques et Cottiennes, Vol. II. p. 600. 
 
 3 Viaggio in Savoia, Lettera LXVI. 
 
\/ 
 
 80 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 81 
 
 would, on their arrival at La Croix d'Aiguebelle, have 
 followed the line of the valley of La Roche tte. The inha- 
 bitants of the Graisivaudan, in like manner, when wishing 
 to penetrate into the Maurienne, would also, when the 
 marshes beyond Pont Charra hindered their further pro- 
 gress up the Isere, have been induced, if not compelled, to 
 strike into the same valley of La Rochette, either from 
 Pont Charra or Le Cheylas. But from La Rochette to 
 Grenoble, the most natural line of route is by AUevard and 
 Le Cheylas. On this line, then, the old Gallic way would 
 have been made ; and here Hannibal would find his road 
 when on his march from Grenoble to the Mont Cenis. 
 
 On this road the marshes present no obstacles. There 
 is a narrow tract of marsh on the plain between Allevard 
 and St Pierre d'Allevard, the highest part of the route. 
 This tract, however, the road can skirt without expe- 
 riencing any difficulty, on the gently rising ground at the 
 foot of the ridge of hills called Brame-Farine. As to the 
 marshes at the extremity of the valley of La Rochette, 
 which extend from near Bettonet to the Arc, the road, 
 being already on the ground lying on their south side, 
 cannot, of course, be inconvenienced by them. 
 
 From the entrance to the valley of the Arc, at La 
 Croix d'Aiguebelle, to the summit of the Mont Cenis, the 
 way, lying in a valley confined on both sides by mountains, 
 necessarily follows up the Arc, as closely as the difficulties 
 of the ground will allow it. Four lines of way up the 
 Maurienne may be mentioned ; (1) The present high road ; 
 (2) the first carriage-road made under Napoleon ; (3) the 
 mule-road of the last century ; and (4) the earlier mule- 
 road, the oldest path known to have existed. These roads 
 are, in places, more or less coincident with one another : 
 the first and the last have, however, an exceedingly small 
 number of miles in common. For the ancient road gene- 
 rally ran, and sometimes at a great height, along the flanks 
 of the mountains, avoiding the marshes in the Lower 
 Maurienne, and the deep defiles in the Upper Maurienne. 
 An artificial line of causeway now carries the high road 
 straight through the marshes ; and the defiles of the 
 Upper Maurienne have been laid open for its passage, 
 
 / 
 
 either by blasting away the rocks which close in upon the 
 river, or by means of artificial terraces which deprive it of 
 a part of its channel. 
 
 It is necessarily along the most ancient path, the 
 earliest mule-road, that we must seek for the route of 
 Hannibal. Now, though this road is partially destroyed in 
 several places, the whole line may be almost completely 
 traced. From La Croix d'Aiguebelle, mentioned above, it 
 skirts the bases of the mountains to Aiguebelle. Directly 
 behind Aiguebelle, two hills rise from the bed of the valley, 
 leaving between themselves and the mountains on each 
 side only three narrow openings. Through the most 
 easterly of these openings runs the high road, supported 
 on a causeway occupying part of the old channel of the 
 river, which it follows to the bridge of Argentine. The 
 first carriage-road and second mule-road passed through 
 the opening between the two hills ; the oldest path of all 
 passed through the opening between the western hill and 
 the mountains, and fell into the line of the first carriage- 
 road and second mule-road immediately behind the hill. 
 This first carriage-road joins the present one shortly before 
 reaching the bridge of Argentine, the junction taking 
 place at the foot of a precipice, which in ancient times 
 must have sprung directly from the river or the marshes, 
 and, in all probability, have barred this line of way. The 
 oldest road is thus, it may be supposed, if there was no 
 passage here, represented for a short distance by a mule- 
 road, which, after the hills previously mentioned behind 
 Aiguebelle have been passed, strikes off from the second 
 mule-road, into which the first has just fallen, and ascends 
 along the heights towards the mines of St George. From 
 this road, in a short time, a foot-path branches off to the 
 left, and, keeping for some distance along the sides of the 
 mountains, eventually descends through a hollow directly 
 upon the bridge of Argentine. This, it may be conjectured, 
 is the oldest route through the hills and defile of Aiguebelle. 
 
 From the bridge of Argentine the old line of way is 
 known throughout by the inhabitants of the country as far 
 as the summit of the Mont Cenis. It passes through 
 Aro-entine to the left of the marshes and of the modern 
 
 6 
 
82 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 carriage-roads (though Argentine itself is accessible in a 
 carriage) and encounters no difficult ground until it ap- 
 proaches Epierre. Here there is a spur projecting from 
 the mountains, and presenting towards the west a precipi- 
 tous face, which rises from the marshes. On the crest of 
 this spur is a tract of cultivated ground. Over this tract 
 the old road passed, attaining it by a gradual ascent along 
 the flanks of the mountains which stretch towards Argen- 
 tine. On the northern side of this tract of ground the 
 road has been destroyed, and its line is lost for a short 
 time among the fields. On the summit of the spur it is 
 again recovered, and descends in a well-preserved state 
 upon Epierre, passing through the most ancient part of 
 this village, where the church is situated. Continuing its 
 course through Le Tardy it crosses another, but a lower 
 spur of the mountains, and descends upon Tigny. From 
 thence, having passed through La Chapelle, it quits, near 
 Gondran, the large tract of fertile ground which extends 
 into the marshes, and upon which Tigny, La Chapelle, and 
 Gondran are situated ; and then pursues its course along 
 the mountain sides through Les Chavannes, till it arrives 
 immediately behind the post-station of La Grande Maison\ 
 Erom thence it reaches La Chambre, keeping along the 
 side of a steep mountain, and coinciding probably with the 
 earliest carriage-road now rapidly falling into complete 
 ruin. Erom La Chambre it passes through St Avre to the 
 bridge of Sainte Marie de Cuines, which is situated about 
 a mile above the village of that name, at a point where a 
 projecting edge of rocks formerly obliged the road to cross 
 the river. It does not however seem unlikely, from the 
 nature of the country, that the river may have been once 
 crossed a little below this bridge, so that the way from La 
 Chambre may have passed through Sainte Marie itself At 
 the bridge of Sainte Marie de Cuines, the great marshes 
 of the Lower Maurienne may be said to terminate. A 
 long defile bounded by precipices of vast height is then 
 
 1 From La Grande Maison to St Martin do la Porte, a distance of 
 about 15 English miles, the first three roads up the Maurienne deviate 
 little from each other. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 83 
 
 entered. The river is again crossed at Pontamafrey, and 
 also a third time at the upper end of the defile. The 
 valley then opens out, and the road emerges upon an open 
 tract of country in which stands St Jean de Maurienne, 
 the chief town of the province of the same name. 
 
 From St Jean de Maurienne to St Martin de la Porte, 
 the way lies in the bottom of the valley, which has here 
 for a time a considerable width. The river is again crossed 
 between St Jean and St Julien. The present high road 
 keeps close to the river, and leaves the villages of St Julien 
 and St Martin on the left. Almost immediately beyond 
 St Martin the valley is suddenly terminated. The great 
 rock of Baune, several hundred feet in height, and about 
 three quarters of a mile in length, stretches across the 
 valley from side to side. Between the southern edge of 
 this rock and the precipices on the opposite side of the 
 valley, the Arc has cut itself a narrow passage between 
 the clifi's. Through this passage the three most modern 
 roads were conducted, the face of the precipice having 
 been blasted away by the aid of gunpow der to make room 
 for them. An entrance was thus gained into the little 
 plain of St Michel, and into a series of defiles about ten 
 miles in length, which extends as far as a village called 
 Le Frene^^ some three miles below Modane. 
 
 No such course was practicable for the ancient road ; 
 nor, according to the report of the inhabitants, did any 
 way exist, till within the last 150 or 200 years, by the 
 southern extremity of the Rock of Baune. The old road 
 passed by the northern extremity of that rock, penetrating 
 ' through the narrow defile of La Porte, which separates 
 the Rock of Baune from the rest of the ridge, of which it 
 forms the prolongation into the valley. Having passed 
 through this defile the road did not descend upon the 
 plain of St Michel (in ancient times probably a morass 0, 
 but climbed the heights by Le Village de la Porte and 
 Villard Bernond to a village called La Traverse; from 
 whence it ran, nearly on a level, round a projecting moun- 
 
 s 
 
 1 According to Bertolotti {Viaggio in Savoia, Lettera LXX.) perhaps 
 
 a lake. 
 
 6 — 2 
 
 k 
 
 * I 
 
 ^i 
 
8^ 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 tain-shoiilder, to Le Thyl d'en haut, the highest point of 
 this part of the route, and probably about 1500 feet above 
 the valley beneath. Between Le Village de la Porte and 
 Yillard Bernond there is only a foot-path now; and be- 
 tween Villard Bernond and La Traverse a considerable 
 portion of the former road is gone, though fragments of 
 the old rough paving sometimes appear. The rest of the 
 way from St Martin de la Porte to Le Thyl d'en haut 
 is good. From Le Thyl d'en haut, the old road rapidly 
 descends upon Le Thyl d'en bas and La Buffaz ; from 
 whence, still descending, but less rapidly, along the moun- 
 tain side through Basilieres, it arrives on the banks of a 
 small tributary of the Arc, having almost reached the level 
 of that river. From Le Thyl to this little tributary stream 
 the road is good ; but, after the stream has been crossed, 
 nothing but a narrow and bad foot-path remains to indi- 
 cate the course of the old road. Before reaching Orelle 
 the road again becomes wide and good, the old track being 
 here followed by the present mule-road which branches 
 off to Orelle from the high road in the defile beneath. 
 From Orelle to Le Freney the old road is still in use and 
 in very good preservation. It runs at a great height along 
 the mountains through the villages of Bonvillard and Le 
 Villaret^ From Le Villaret it descends upon St Andre, 
 through which the later mule-road and the first carriage- 
 road also passed ; and another descent leads down to the 
 bridge of Le Freney, where all the lines of road again 
 meet, and where an open valley is once more reached-. 
 
 1 In a detached part of this latter village, called, as well as the sound 
 could be caught, La Paille, there is an old house with early Gothic 
 windows, which is reported by the inhabitants of the place to have been 
 formerly a post-house for mules or horses, when the old road passed 
 this way. 
 
 2 The course of this old road, from St Martin to Le Freney, may be 
 frequently discerned from the high road by the side of the Arc beneath. 
 Between La Buflfaz and Basilieres it runs in one place along the face of a 
 precipitous mountain where its actual line is perceptible. A similar 
 place occurs between Bonvillard and Le Villaret. Where the line of road 
 itself cannot be seen, its course may be partially traced by means of the 
 numerous crosses along it, and by the villages, when visible, through 
 which it runs. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 85 
 
 From Le Freney to Modane the valley is open, especially 
 around Modane. The old road passes through the villages 
 of Le Freney and Le Fourneau, to the right of the present 
 high road. Above Modane the defile of Bramans or of 
 1 Esseillon, a passage of great difficulty, is entered. The 
 high road (the two carriage-roads are here identical) is cut 
 out of the rock for the greater part of the way to the 
 plain of Bramans. There seems also never to have been 
 more than one line here for mules or horses. Leaving 
 Modane, this track passes to the right of the carriage-road, 
 being separated from it by an eminence, behind which the 
 older road makes its way. Emerging from behind this 
 eminence, it descends to, and cuts across the high road, 
 and leads down to Villarodin. Mounting again from this 
 village, and re-crossing the high road, it ascends the 
 heights into a wood, rejoining after a time, but only for 
 a moment, the high road, at the bridge which crosses a 
 small stream opposite the great fort of I'Esseillon. The 
 mule-road then again ascends the heights, and is lost 
 in the pine-forests, and finally descends upon Bramans 
 behind a ridge of some elevation, which intervenes between 
 it and the carriage-road. 
 
 On the plain of Bramans the roads to the Great and 
 Little Mont Cenis separate. The way to the Little Mont 
 Cenis, the pass which agrees best with the accounts of 
 Hannibal's passage, and which seems to have been the 
 most ancient route by which the mountain was ascended S 
 here strikes off from the valley of the Arc, and ascends the 
 lateral valley of St Pierre. The old mule-road to the 
 Great Mont Cenis passed from Bramans through Le 
 Verney, Le Chatel, and Sollieres, to Termignon; from 
 whence, with the exception of a small portion on leaving 
 Termignon, where the carriage-road makes a zigzag, it nearly 
 coincided with the line of that road as far as Lanslebourg, 
 a large village at the foot of the Great Mont Cenis. From 
 Lanslebourg it forms what is called the * Petite Route' by 
 which the mountain is ascended. The high road is only 
 
 ^ Histoire du diocese de Maurienne, par M. Lo Chanoine d'Angley. See 
 also below. Chap. XL 
 
86 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 able to accomplish this ascent by means of several zigzags. 
 At La Ramasse, the northern extremity of the plateau of 
 the Great Mont Cenis, the two roads unite, and follow 
 nearly the same line to La Grande Croix, at the southern 
 extremity of the plateau. This plateau is five miles long. 
 The highest point of the passage, 6770 feet above the level 
 of the sea, is situated about three-quarters of a mile to the 
 south of La Eamasse. At La Grande Croix the old road 
 from the Little Mont Cenis falls into the present line 
 of routed 
 
 The road to the Little Mont Cenis has been already 
 mentioned as diverging from the valley of the Arc at 
 Bramans. It strikes thence up the valley of St Pierre, 
 and, after a course of about five miles, reaches the foot of 
 the Little Mont Cenis. A very steep ascent, of one hour 
 in length, leads to the crest of the pass, 6820 feet above 
 the sea-level. The summit of the Little Mont Cenis is a 
 plateau of grass, of considerable width, and exceeding 
 three miles in length. It is enclosed by mountains on both 
 sides, and terminates towards the N. E. in a slope, which 
 declines to the lake of the Mont Cenis. The path, on 
 arriving at this opening above the lake, does not descend 
 to it, but turns off towards the south-east, and enters a 
 kind of short valley, having the mountains on the right, 
 and on the left a range of hills, mentioned in the note as 
 forming on the south-west the boundary of the plateau of 
 
 1 There are three paths leading from the Little Mont Cenis to the 
 high road which runs along the plateau of the Great Mont Cenis. The first 
 lies along the northern side of the lake, and falls into the high road at 
 Les Tavernettes. The second runs to the Hospice by the southern side 
 of the lake, along the extremity of a range of hills which forms the 
 boundary of the plateau of the Great Mont Cenis, from the lake to La 
 Grande Croix. The third crosses these hills through a hollow in them, 
 and descends directly upon La Grande Croix. This last, the shortest and 
 most natural way, is the ancient lino from the Little Mont Cenis into 
 Italy. In the map given in Blaev's Theatnmi Statuum Regioe Celsitudinis 
 Sabaudice DuciSt the roads from Italy over the Great and the Little 
 Mont Cenis are marked as separating at La Grande Croix or even a little 
 before. The road to the Little Mont Cenis seems to have skirted the 
 base of the hills, and to have left the present hamlet of La Grande Croix 
 some hundred yards or so to the right. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 87 
 
 the Great Mont Cenis. Through a hollow in these hills, as 
 there stated, the path finally passes, and rejoins, at La 
 Grande Croix, the route of the Great Mont Cenis. The 
 descent into Italy, which begins at La Grande Croix, is 
 the same for both passes. 
 
 This descent is at first very steep. The road winds 
 down a precipitous declivity, called the ' Fourmiguier,' by 
 which the little plain of St Nicholas, lying perhaps about 
 500 feet below La Grande Croix, is at length reached. The 
 face of this declivity is traversed from side to side by the 
 several arms of the high road, which is cut in places out 
 of the rock, and is only able to accomplish the descent by 
 a zigzag route. By these arms of the high-road, rising 
 one above another, the old mule-road seems to have been 
 quite obliterated. The little plain of St Nicholas, once 
 reached, is soon crossed. At the extremity of this plain 
 the carriage-road (always the same from this point), sepa- 
 rates from the common line of the old mule-roads, and is 
 only joined by it again at Susa ; the extreme steepness of 
 the descent to La Novalese rendering that line of route 
 very ill adapted for a carriage-road. The first part of the 
 descent from the plain of St Nicholas, by the mule-road, 
 lay through a covered gallery, built for the protection of 
 travellers, and now in a great measure buried beneath the 
 fragments of rock which were detached from above, when 
 the mountain-side was blasted away to afford a passage for 
 the high-road. The rest of the descent, as far as La 
 Ferriere, is rapid, but not excessively steep. At La Ferriere 
 the first and second mule-roads separate, and unite again 
 at La Novalese. The second road, strongly constructed, 
 and paved with large stones, was made under Charles 
 Emmanuel HI. (1730 — 1773) in consequence of the dan- 
 gers which attended the oldest line of way. This second 
 mule-path from La Ferriere keeps on the right or southern 
 side of the Cenise or Cenischia : the old path lay on the 
 northern side of the stream. The descent between La 
 Ferriere and La Novalese is very rapid, and in parts pre- 
 cipitous. 
 
 ^ At La Novalese the descent from the Mont Cenis 
 is terminated. This village, though so near the summit of 
 
88 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 the pass, lies at an elevation of no more than 2560 feet 
 above the sea. From thence to the plains of Italy the 
 route presents no further difl&culty, unless the passage of 
 a low ridge, immediately before reaching Susa, be con- 
 sidered as one. Yet even this obstacle might, it should 
 seem, have been avoided, if necessary, by following the 
 Cenise from Venaus, on its left bank, down to its junction 
 with the Dora Susina in the valley, or Combe, of Susa. 
 From La Novalese downwards, there is no further question 
 of mere mule-roads, as the country does not offer any im- 
 pediment to the construction of carriage-roads. In such 
 a district it is not of much importance to determine the 
 oldest route. From Susa to Bussoleno, on the way to 
 Turin, the northern side of the Dora affords the best line 
 of road, though the southern side may have been anciently 
 followed. From Bussoleno to Turin the southern side has, 
 till late years, always been taken, though the river is now 
 crossed at Borgone, instead of Bussoleno. The earliest 
 road seems to have passed, after leaving Bussoleno, 
 through the towns or villages of San Giorio, Sant' Anto- 
 nino, Vayes, Chiusa, Sant' Ambrogio, Avigliana, Buttig- 
 liera, Rosta, and Rivoli, and thus to have finally reached 
 the city of Turin. 
 
 ,^,. ., .M.afc t -^ 
 
 
 1 . 
 
 x 
 
 V 
 
f 
 
 CHAPTER VII. 
 
 Application of the conditions deduced from Polybius* narrative to the 
 route of the Mont Cenis. — Accordance of the actual distances 
 with those given by Polybius. — Identification of the defile at the 
 commencement of the ascent of the Alps with the gorge of Le 
 ;p^y — Probable identity of the town of the Allobroges and the 
 modem AUevard. — Identification of the " strong white rock" with 
 the rock of Baune. — Arrival at the Col of the Little Mont Cenis, 
 
 THE first two conditions for the determination of Han- 
 nibal's route relate to two distances ; (1) the distance 
 from the junction of the Rhone and Isere, (or from 
 Valence) to the point where Hannibal would have left the 
 Isere ; and (2) the distance from this latter point, across 
 the Alps, to the com-mencement of the plains of Italy. It 
 will therefore be necessary to compute : (1) the length of 
 the road along the left bank of the Isere from Valence to 
 Le Cheylas ; and (2) the length of the Alpine route across 
 the Little Mont Cenis from Le Cheylas to Avigliana. 
 
 Now from Valence to Grenoble the distance is 96 kilo- 
 metres; and from Grenoble to Le Cheylas the distance 
 is 33 kilometres. The total distance, therefore, from 
 Valence (or from the junction of the Rhone and Isere) to 
 Le Cheylas is 129 kilometres or 87^ Roman miles'. 
 
 The following table will give the distances between the 
 different places in succession on the way from Le Cheylas* 
 
 1 A Roman mile is equal to 1476 metres. 
 
 2 The name Cheylas is probably identical with chald>, a word existing 
 in the patois of Dauphine, and signifying a footpath. (" ChaU, petit 
 sentier, parti culierement celui qu'on fait sur la neige." ChampoUion- 
 Figeac, Nouvelles recherches sur les pdtois.) Chald, however, of which an- 
 other form is chal, seems to be a corruption of the Latin scala, and would 
 have signified originally an ascending path, and thence a mountain path 
 generally. *' II est k presumer que la montee de Chalemont (a place by 
 Grenoble) fut ainsi nommee du vieux mot chal, qui signifie route, c'est- 
 k-dire, route de la montagne. Nous avous trouve dans un manuscrit 
 scala montis'^ Pilot, Histoire de Grenoble. In the Val. d'Orca, also. 
 
 .( 
 
CHAPTER VII. 
 
 Application of the conditions deduced from Polybius' narrative to the 
 route of the Mont Cenis. — Accordance of the actual distances 
 with those given by Polybius. — Identification of the defile at the 
 cornmencenient of the ascent of the Alps with the gorge of Le 
 j^^^y — Probable identity of the town of the Allobroges and the 
 modern Allevard. — Identification of the " strong white rock" with 
 the rock of Baune. — Arrival at the Col of the Little Mont Cenis. 
 
 THE first two conditions for the determination of Han- 
 nibal's route relate to two distances ; (1) the distance 
 from the junction of the Rhone and Isere, (or from 
 Valence) to the point where Hannibal would have left the 
 Isere ; and (2) the distance from this latter point, across 
 the Alps, to the commencement of the plains of Italy. It 
 Avill therefore be necessary to compute : (1) the length of 
 the road along the left bank of the Isere from Valence to 
 Le Cheylas ; and (2) the length of the Alpine route across 
 the Little Mont Cenis from Le Cheylas to Avigliana. 
 
 Now from Valence to Grenoble the distance is 96 kilo- 
 metres; and from Grenoble to Le Cheylas the distance 
 is 33 kilometres. The total distance, therefore, from 
 Valence (or from the junction of the Rhone and Isere) to 
 Le Cheylas is 129 kilometres or 87^ Roman miles ^ 
 
 The following table will give the distances between the 
 different places in succession on the way from Le Cheylas^ 
 
 1 A Roman mile is equal to 1475 metres. 
 
 2 The name Cheylas is probably identical with chaU, a word existing 
 in the patois of Dauphine, and signifying a footpath. (" Chala, petit 
 sentier, particulierement celui qu'on fait sur la neige/' Champollion- 
 Figeac, Nouvelles recherches sur les patois.) Chald, however, of which an- 
 other form is chal, seems to be a corruption of the Latin scala, and would 
 have signified originally an ascending path, and thence a mountain path 
 generally. *' H est a presumer que la montee de Chalemont (a place by 
 Grenoble) fut ainsi nommee du vieux mot chaly qui signifie routey c'est- 
 a-dire, route de la montagne. Nous avous trouve dans un manuscrit 
 scala montis" Pilot, Histoire de Grenoble. In the Val. d'Orca, also. 
 
90 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 to Avigliana. The government Itinerary of the Sardinian 
 States, published in 1839, gives the distances on the road 
 then existing between La Eochette and Bramans in Pied- 
 montese miles. On the older road the distances arc of 
 course somewhat different : but, with the exception of the 
 part between St Jean de Maurienne and Modane, the dif- 
 ferences must be slight, and the distances of the Itinerary 
 may, with that exception, be adopted. The distances 
 between Le Cheylas and La Eochette, between St Jean de 
 Maurienne and Modane, and between Bramans and Susa 
 by the Little Mont Cenis, — distances for which no table 
 could be consulted, — have been estimated with consider- 
 able care, principally, but not solely, by observing the time 
 occupied in walking between different points, and may be 
 relied upon as free from all material error. The distance 
 of 24 Eoman miles between Susa and Avigliana (Ad fines) 
 is taken from the Eoman Itineraries. 
 
 A Piedmontese mile is equal to 2466 metres, and a 
 Eoman mile is equal to 1475 metres, or 1614 English yards. 
 Five Eoman miles will consquently be almost accurately 
 equivalent to three Piedmontese miles, and twelve Eoman 
 miles will be equivalent to eleven English miles. 
 
 there is an ascent called the Scalare di Ccresol. The use of the German 
 word, Steg, in the Swiss x\lps, seems to correspond to the use of chald in 
 the Alps of Dauphine. The meaning of Steg is given as simply path, 
 but it is derived from Steigen, to mount, and, under the forms Steg, 
 Steig, or Gsteig, is found in several places where ascents begin. (Ex. 
 Gsteig, Sanetsch pass, N. side, and Simplon pass, S. side; Am-steg, StGo- 
 thard road ; Steg, entrance to Lutsch Thai.) The terms Monta, Montets, 
 Montee, are used in some districts of the Alps where French is spoken 
 to designate similar places. The word ava^oKrj seems quite identical 
 in meaning with such terms as Monta and Gsteig, and with the term 
 Chala in its primitive signification. As there are in the Alps at the 
 present day, so also there would have been in the time of Hannibal, 
 several places known and designated as points where ascents began. 
 All such points might be called ava^okal by a Greek writer. The dva- 
 /SoX)}, of which Polybius speaks, would bo one of these points, and, as 
 is evident from the context, the first point where the road into Italy 
 from the banks of the Rhone ascended into the mountains. Le Cheylas 
 is accurately in this position, and probably derives its name from its 
 situation, as being what Polybius calls the ava^oX^ rav "AXnfav tj fts 
 'iraXtav, or, in another place, the dva^oXrj irphs ray'AXTrftff, 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 91 
 
 From Le Cheylas to Allevard 
 
 Allevard „ La Rochette 
 
 Pied. 
 Miles. 
 
 La Rochette 
 Chamoux 
 Aiffuebelle 
 La Chapelle 
 La Chambre 
 St Jean 
 
 St Martin 
 
 )i 
 
 >f 
 
 59 
 
 95 
 
 95 
 
 95 
 
 55 
 
 i} 
 
 Orelle 
 
 I" 
 
 Orelle 
 
 St Andre 
 
 Modane 
 
 L'Esseillon 
 
 Bramans 
 
 Col 
 
 Grand' Croix „ 
 
 La Novalese „ 
 
 *5 
 
 J> 
 
 99 
 
 95 
 
 59 
 
 95 
 
 Chamoux 
 
 Aiguebelle 
 
 La Chapelle 
 
 La Chambre 
 
 St Jean de Maurienne ... 
 St Martin de la Porte . . . 
 
 to Villard Bemond U 
 
 La Traverse H 
 
 BasilJeres 2 
 
 Orelle U' 
 
 St Andre 
 
 Modane 
 
 L'Esseillon 
 
 Bramans 
 
 Col of Little Mont Cenis 
 
 Grand' Croix 
 
 La Novalese 
 
 Susa 
 
 4i 
 4,1 
 
 English 
 Miles. 
 
 61 
 
 3 
 2 
 
 6 
 
 si 
 
 Roman 
 Miles. 
 
 ~6~ 
 
 7 
 
 '2 
 
 7 
 6i 
 
 72 
 '3 
 
 61 
 6 
 
 5 
 
 Si 
 Si 
 
 Si 
 
 71 
 
 6i 
 
 H 
 24 
 
 132i 
 
 Susa to Avigliana or " Ad Fines" 
 
 Total distance from Le Cheylas to Avigliana 
 
 The first two conditions are these : 
 
 I The commencement of the ascent of the Alps must 
 be situated at a distance of about 100 Roman miles from 
 the junction of the Rhone and the Isere, reckoned along 
 the left bank of the latter river. ^ ^ 
 
 II. The length of the route over the Alps, begmmng 
 at the commencement of the ascent of the mountains, and 
 terminating at the commencement of the plains of Italy, 
 must be about 150 Roman miles. 
 
 Instead of the 100 Roman miles of (L) the route gives 87^ 
 
 150 „ „ (II.) » " 1^-2 
 
 The actual distances thus fall, in both cases, a little 
 below those given by Polybius. This is exactly what ought 
 to occur when the distances are estimated, as they are by 
 Polybius, in stadia : for the estimated distances in stadia 
 always exceed the real distances^. It has also been pre- 
 
 1 « The principal argument for a variety of stadia is that of Major 
 Rennell, iGeog. of Herod, s. 2.); namely, that when ancient authors have 
 
92 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 viously noticed, that Polybius measures distances by lengths 
 of 200 stadia, or 25 Eoman miles. Now 100 is the lowest 
 multiple of 25 which exceeds 87^; and 150 is also the 
 lowest multiple of 25 which exceeds 132^. The argument, 
 therefore, drawn from the comparison of distances, could 
 not be more favourable than it is to the route of the Little 
 Mont Cenis. 
 
 We now proceed to the consideration of Condition III. 
 
 III. At the commencement of the ascent of the Alps, 
 a defile must be found, the character of which, and of the 
 surrounding country, is in accordance with the events 
 related by Polybius to have occurred in that place. 
 
 The chain of lofty hills, which bounds the valley of 
 Graisivaudan on the east, has been already mentioned. 
 At Le Cheylas, there is a gap in this chain, affording a 
 passage to a small stream, called the Torrent des Sailhes, 
 which rises in the elevated valley behind the hills. This 
 gap or defile is called the gorge of Le Fay. It is com- 
 pressed on the north by the extremity of the ridge of 
 Brame-Farine, and on the south by a very elevated hill 
 crowned by five sharp summits, and bearing, on that 
 account, the name of Les Cinq Pointes. Through the 
 gorge of Le Fay, which divides these hills, admission is 
 gained from the Graisivaudan into the valley of Allevard. 
 
 The road from Allevard to St Pierre d' Allevard lies 
 over a plain, enclosed between the ridge of Brame-Farine 
 and the higher Alps. This plain is marshy in the middle : 
 the road leaves the marsh on the east, and skirts the steep 
 slopes of Brame-Farine on the west. At its southern ex- 
 tremity the ridge of Brame-Farine forks, as will be seen 
 by the plan, into two spurs. From St Pierre d'Allevard, 
 the road to Le Cheylas begins to descend, and winds round 
 the base of the eastern spur of Brame-Farine. The descent 
 eventually becomes rugged, and the base of the hill, * Les 
 Cinq Pointes,' begins to approach the road. The old path 
 
 stated the distances between known places, and a comparison is made 
 between their statements and the actual distances, the distances stated by 
 them are invariably found to be too great, never to small" — Penny Cyc. 
 Art. " Stadium," which see. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 93 
 
 here lies immediately to the left of the carriage-road, which 
 is partly cut out of the rock. After the bridge over the 
 Torrent des Sailhes is reached, the defile becomes very 
 contracted. The modern road here crosses the torrent by 
 a bridge, and pursues its course to Goncelin, the rocks 
 having been blasted to gain a passage for it. After passing 
 under this bridge the torrent begins to sink in a rift, and 
 is soon lost in a deep and narrow trench, bounded on both 
 sides by sheer precipices, between which it flows for a full 
 mile, until it emerges into the Graisivaudan at Le Cheylas. 
 Two roads branch off, along the extremity of the western 
 spur of Brame-Farine, from the bridge over the torrent. 
 The first and most direct is a footpath, which keeps along 
 the edge of the precipices ; the second, a cart-road, con- 
 structed higher up the face of the declivity, and avoiding, 
 at the expense of a slight ascent and subsequent descent, 
 the dangers of the precipices. The foot-path, the shortest 
 and most natural way, would be the earliest road. For 
 about a quarter of a mile from the bridge it lies over 
 a small tract of nearly level ground, crossing a little 
 stream on its way. The projecting edge of the western 
 spur of Brame-Farine then forces it to the edge of the 
 precipices, and for half a mile it is obliged to skirt them 
 closely, and frequently to run along their very brink. At 
 length it is enabled to turn away from this dangerous 
 vicinity, and to descend, by a long and winding course, 
 down the western slopes of Brame-Farine, to Le Cheylas. 
 
 Here, then, in the gorge of Le Fay, we find a defile 
 existing at the first commencement of the ascent to the 
 Alps. Its character also, and the nature of its vicinity, 
 will prove, upon examination, to be in accordance with the 
 requisitions of Polybius' narrative. For 
 
 (1) The defile must be commanded by certain heights 
 of considerable elevation. 
 
 These heights are those of Brame-Farine'. The gene- 
 ral character of this ridge is that of an exceedingly steep 
 slope, partly covered with brush-wood, and partly with 
 
 1 The heights commanding the defile are called by Livy " tumuli," 
 (i. e. hills as distinguished from mountains.) It will be seen that the 
 
 heights of Brame-Farine are of this nature. 
 
94 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 grass. Some more level tracts than usual are, however, 
 under cultivation. The ridge, as previously mentioned, 
 forks, at its southern extremity, (marked D on the plan) 
 into two spurs. From Z>, the ridge-line of the eastern spur 
 descends to a, and then expands into a small level tract 
 Ai the most commanding portion on the heights. The 
 ridge-line of the western spur also eventually expands into 
 a similar tract B ; but this tract is considerably lower than 
 Ay and is commanded by it. A rapid descent intervenes 
 between the tract B, and another piece of level ground (7, 
 on which stands* a group of cottages. From C, short steep 
 slopes descend to the foot-path running along the edge of 
 the precipices beneath. 
 
 These heights are accessible hy various tracks. One 
 of them is traced upon the plan. By this path it is a walk 
 of about three quarters of an hour, up a steep ascent, from 
 St Pierre d'Allevard to the point a, the northern extremity 
 of the level tract A, From a it is a descent of nearly half 
 an hour, along the sides of rapid slopes, to 6, on the edge 
 of the tract B. From h to the group of cottages is another 
 descent of about twenty minutes. The old foot-path 
 through the defile lies at a short distance below this group 
 of cottages. 
 
 From these measurements of time it will be seen, that 
 the heights A and B, commanding the defile, are of con- 
 siderable elevation. The level tract Aj the most important 
 of the two positions A and By can, indeed, hardly be less 
 than 1500 feet above the Graisivaudan. 
 
 (2) These heights must not be easily accessible from 
 below the defile. 
 
 From the side of the Graisivaudan these heights 
 are certainly not easily accessible, especially the more 
 commanding of the two posts, that marked A, But from 
 the valley of Allevard this post is much more readily 
 gained, both as being nearer, and as requiring an ascent, 
 of which the perpendicular elevation is considerably less. 
 
 (3) The way through the defile must skirt in some 
 place the edge of a precipice. 
 
 It has been already mentioned, that the foot-path from 
 Le Cheylas through the gorge of Le Fay skirts the edge of 
 a precipice for a distance of about half a mile. 
 
 , 
 
 
 
 f 
 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 95 
 
 (4) Immediately below the defile must be found a 
 place where an army as numerous as that of Hannibal 
 could encamp. 
 
 The Graisivaudan, from Grenoble to Pont Charra, is a 
 broad and level vale throughout. On the ground between 
 Le Cheylas and Goncelin, Hannibal might encamp without 
 
 difficulty. 
 
 (5) This place of encampment should be visible from 
 
 the heights commanding the defile. 
 
 From the heights A, a large part of the country be- 
 tween Goncelin and Le Cheylas is visible, as well as the 
 western part of the Graisivaudan down to Grenoble. From 
 the position B the rest of the eastern side of the Graisi- 
 vaudan is visible in addition. 
 
 (6) Above the defile, and near it, must be found an 
 open district, where a town either stands, or might have 
 stood, and where an army such as that of HannibaPs could 
 have encamped. 
 
 Such an open district is the valley of Allevard. Its 
 length, reckoning from St Pierre d'Allevard to the Savoy- 
 ard frontier, is nearly six miles. Its breadth is about half 
 an hour's walk from side to side^ The town which Han- 
 nibal took would probably be where Allevard, the chief 
 place in the district, now stands. Polybius calls this town 
 7ro\t9, which would lead us to infer that it was a place of 
 some importance in the country of the Allobroges, where 
 Hannibal then was. Livy mentions it as being the capital 
 of the district. It would therefore either be the chief 
 town of the Allobroges, then perhaps a smaller tribe than 
 that which afterwards bore this name, or else the chief 
 town of one of the districts into which their country was 
 divided, the capital of one of the smaller tribes which 
 together composed the whole people. That the chief town 
 of an AUobrogic district should have been situated at 
 Allevard seems probable, both from the present importance 
 of the place, and from its appearing to preserve, in its own 
 name, that of the Allobroges^ 
 
 1 Guide du Visiteur an pays d^ Allevard. 
 
 2 The root of the latter part of the name of the Allobroges, called 
 
96 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 97 
 
 Having now shewn the accordance of the character of 
 the gorge of Le Fay, and of its neighbourhood, with the 
 requirements of Polybius* narrative, it may be advisable, 
 before proceeding further, to obviate an objection which 
 might be raised against considering Le Cheylas as lying 
 at the commencement of the ascent of the Alps. It might 
 be said that as, from the extremity of the Graisivaudan, 
 near Voreppe, a point about 30 miles below Le Cheylas, the 
 road has lain through a valley bounded on both sides by 
 mountains, the commencement of the ascent of the Alps, 
 il TT/oJs ras'iWTreis ava^okn. ought to be fixed there. Yet 
 * the commencement of the ascent of the Alps' need not 
 signify the geographical commencement of that mountain- 
 system, but the point where the route first begins to ascend 
 the mountains. Indeed, the term is much more applicable 
 to a locality of the latter kind. Now the route up the 
 wide and level vale of Graisivaudan has nothing of a moun- 
 tainous nature. It is at Le Cheylas, where Hannibal is 
 supposed to have left the Isere, and turned off to AUevard, 
 that the character of the way changes abruptly from level 
 to mountainous ; so that the commencement of the ascent 
 is here definitely marked. Besides, the mountains on the 
 right bank of the Isere are not properly the Alps, but the 
 Chartreuse mountains, a group perfectly isolated from the 
 Alps by the valley of Chambery and the Graisivaudan. 
 Neither does Polybius seem to have considered them as a 
 part of the Alps ; for, having occasion to mention them in 
 his description of the Island, he speaks of them indefinitely 
 as opt], and as if they were unconnected with the Alps and 
 the Alpine route. In fact, the mountains to the east 
 or south-east of the valley of Allevard are at present 
 
 AUobriges by Polybius, is the frequently recurring Celtic word hrig, 
 which, under the forms hriga, or hrka^ appears as the termination of the 
 names of a great number of ancient towns in Gaul and Spain. That 
 Allevard should be a corruption of AUobriga or Allobrica, or some 
 similar word, is not improbable. The ancient name of Brianyon, Brig- 
 antio, or Brig-antiurriy is found in Ammianus Marcellinus as Virg-antia, 
 or Verg-antium. The name of the ancient Tala-hrica, also, has become 
 corrupted into the modern Tala-vera; and the ancient Baudo-brica, (or 
 Bonto-hrica) on the Rhine, is now identified with the modern Bop part. 
 
 \\ 
 
 spoken of as being the 'first' of the Alps, *les premieres 
 montagnes alpines^ ;' so that when Hannibal ascended, by 
 the gorge of Le Fay, to the valley of Allevard, througli the 
 chain of hills intervening between that valley and the Grai- 
 sivaudan, it might be said of him, with perfect accuracy, 
 that he then i^P^^^'^o t^s irpos rd^ "A\7rets di/a/3oX^9. 
 
 When Hannibal first entered, or approached, the country 
 of the AUobroges, they were afraid of attacking him, 
 partly on account of his cavalry, and partly on account 
 of the escort of the men of the Island, who guarded the 
 rear of the Carthaginians. Now the vale of Graisivaudan, 
 from Grenoble to Le Cheylas, being perfectly level, is 
 quite adapted throughout to the action of cavalry, so that 
 the AUobroges would never venture to descend from the 
 hills, and fall upon Hannibal's line of march. The ambus- 
 cade they projected in the defile leading to Allevard was a 
 surer method of destroying the Carthaginians. From the 
 difficulties of the road, and the manner in which it winds, 
 throughout the whole distance from Le Cheylas to Alle- 
 vard, either up, or along the base of the acclivities of 
 Brame-Farine, with steep slopes, precipices, rugged ground, 
 or marshes on its other side; — on account of these disad- 
 vantages, and the centrical position and unassailable nature 
 of the commanding heights of Brame-Farine, an army, 
 marching from Le Cheylas to Allevard, would have been 
 brought into imminent danger of destruction by an attack 
 from the heights, every point in the long line of march 
 being exposed to assault. 
 
 Such a destructive attack was projected by the AUo- 
 broges : but Hannibal had become aware of their plans, 
 and took measures to frustrate them. On the evening of 
 the eighth day after leaving the confluence of the Rhone 
 and the Isere, the Carthaginians, having advanced from 
 their previous position, in full view of the AUobroges on 
 the heights, encamped on the plain between Le Cheylas 
 and Goncelin, before the entrance of the defile. As night 
 drew on, the camp-fires were lighted, and were observed 
 
 ^ Guide du Visiteur au pays d* Allevard, 
 
T7 
 
 98 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 99 
 
 by the Allobroges from their posts on the hills, who, con- 
 cluding that no further movement was meditated by Han- 
 nibal, withdrew from the heights to the neighbouring town 
 of Allevard. But Hannibal, previously aware that the 
 Gauls would withdraw, selected a body of active and lightly- 
 equipped men, passed through the gorge, along the edge 
 of the precipices, in the dead of night, and seized upon 
 the heights which the Allobroges had abandoned. Han- 
 nibal himself would probably choose for his own post the 
 eastern heights', from whence Allevard is seen, and the 
 most comprehensive view of the route obtained. Having 
 thus gained possession of the heights, the Carthaginians 
 took up their position upon them, and waited for the day. 
 
 When morning dawned, the Allobroges prepared to 
 return from Allevard to their former posts, but were 
 seized with dismay for a time, when they perceived that 
 they had been occupied during the night by the Cartha- 
 ginians. Nothing, however, would have prevented the 
 Gauls from advancing towards the defile, and watching the 
 progress of events. They might, without incurring much 
 danger of molestation on the part of Hannibal, have 
 occupied the ground to the south-east and south of the 
 road, from St Pierre d' Allevard to Les Cinq Pointes, or 
 even have advanced along the lower slopes of the eastern 
 spur of Brame-Farine, and have collected to the north of 
 the road leading through the defile. For Hannibal, be- 
 sides being at a considerable distance from the Allobroges, 
 could not fall upon them from above without leaving the 
 heights open to recapture from the side of Allevard. 
 
 At length the leading columns of the Carthaginian 
 army, the cavalry and baggage-animals, began to appear, 
 making their way in a long line through the defile, and 
 much embarrassed by the difficulties of the ground. The 
 Allobroges were unable to resist the opportunity which 
 the defenceless state of their enemy seemed to afford 
 them, and fell upon the line of march at many points. By 
 these attacks the Carthaginians sustained severe losses. 
 A number of horses and baggage-animals, as well as the 
 
 1 Those marked A in the Plan. 
 
 
 men with them, fell into the hands of the Allobroges, and 
 were carried off to their town. Nearly all, probably, be- 
 came captive, who had passed the line of precipices, and 
 were ascending the rugged ground towards St Pierre 
 d' Allevard. Attracted by the hopes of spoil, the greater 
 part of the inhabitants of the town issued out to join in 
 the attack, and the town itself was left with comparatively 
 
 few occupants. . 
 
 The foremost columns of the Carthaginian army havmg 
 been thus cut off, the Allobroges attempted to destroy the 
 rest in detail, as they made their way along the edge of 
 the precipices. Taking possession of a small plateau imme- 
 diately above this part of the roadS they directed their 
 attacks upon the Carthaginians beneath. Numbers of the 
 baggage-animals were thus destroyed, rolling with their 
 burdens down the precipices by the side of the road. The 
 loss was much increased by the horses of the cavalry, 
 which, becoming unruly in consequence of the wounds 
 they received, created great confusion in the line of march, 
 and threw the whole column into disarray. 
 
 Hannibal saw from above the gradual destruction of 
 his army, and determined to charge down from the heights 
 upon the Allobroges, and attempt to preserve the re- 
 mainder of his baggage, on which the subsequent safety of 
 the army depended. Accordingly, accompanied by the 
 men who had seized the heights in the night, he descended 
 from above, and fell upon the enemy^ Attacked at a dis- 
 advantage, in consequence of Hannibal's charging them 
 from the heights, the Allobroges were defeated, and 
 obliged to give way. Borne back towards the precipices, 
 they were thrown upon the Carthaginian column of march 
 beneath them, and involved them in their own fate. The 
 success of Hannibal was thus inevitably attended with the 
 destruction of part of his own army. By no other means. 
 
 1 Marked C in the Plan. , , , , 
 
 2 The group of cottages on the plateau marked C would be the point 
 about which the contest was hottest. Hannibal would probably descend 
 in a direction represented by the footpath marked on the Plan along 
 the heights. ^ 
 
100 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 however, could the safety of the remainder have been 
 assured, and the sacrifice was therefore made. The greater 
 part of the Allobroges were either slain by the sword, or 
 driven down the precipices : the few survivors of the con- 
 flict turned and fled to their own town, or dispersed them- 
 selves among the surrounding mountains. 
 
 Hannibal, however, gave his enemies no time to recover 
 themselves. No sooner was the battle terminated, than he 
 collected together as many of his scattered troops as he 
 could, and marched directly upon the town of the Allo- 
 broges, while the rest of his army was making its way 
 through the defile. The town was enabled to ofler no 
 efi^ectual resistance, (for it was very nearly empty) and fell 
 at once into the hands of Hannibal. By its capture he 
 recovered possession of his own horses, baggage-animals, 
 and men, who had been carried off" by the Allobroges at 
 the commencement of the contest, and obtained a supply 
 of provisions for his army, sufficient for two or three days' 
 consumption. The fugitive Allobroges also, dispersed in 
 the neighbouring districts, diffused such terror among the 
 inhabitants by the story of their defeat, as to make them 
 very cautious of molesting the Carthaginians. AVhen Han- 
 nibal was next attacked, it was not by open hostility, but 
 by a previously concerted system of treachery K 
 
 The fourth condition which has been deduced from the 
 history of Polybius, is the following : 
 
 IV. At a place nearly half-way, in point of time, 
 between the town of the Allobroges and the summit of 
 the pass, a certain 'strong white rock* must be found; 
 and the adjacent country must be in accordance with the 
 events which are recorded by Polybius to have taken 
 place in the neighbourhood of this rock. 
 
 In point of distance, St Jean de Maurienne lies half- 
 way between AUevard and the lake on the summit of the 
 
 1 A good general yicw of the supposed scene of the contest with the 
 Allobroges is gained from the modern road between Goncelin and AUe- 
 vard, shortly before it arrives at the bridge over the Torrent des Sailhcs, 
 on its way from Goncelin. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 101 
 
 Mont Cenis, being about 40 English miles from each. 
 In point of time, however, the half-way point must be 
 placed somewhat above St Jean, as the difficulties of the 
 way are much greater above than below, and Hannibal 
 would have marched most quickly in the Lower Maurienne 
 and the valley of La Eochette, where, in addition to the 
 comparative easiness of the ground, his march was also 
 unimoeded by any aggression on the part of the inha- 
 bitants. The 'strong white rock,' consequently, being 
 nearly, but not quite, half-way, in point of time, between 
 the town and the summit of the pass, must be sought, 
 either in the immediate neighbourhood of St Jean de 
 Maurienne, or a little above. 
 
 It is found accordingly between six and seven miles 
 above St Jean, immediately beyond St Martin de la 
 Torte, being, in fact, the great rock of Baune, already 
 mentioned, (p. 83) as putting a sudden termination to the 
 open valley extending from St Jean to St Martin. This 
 rock seems to possess the characteristics of the Xet/zcoTre- 
 Tpov o-^^vpdi' of Polybius, that is to say, to form a pro- 
 minent feature in the country, and to be remarkable for 
 its whiteness and (military) strength'. 
 
 The rock of 13aune is first seen, from a distance of 
 about seven miles, on the approach to St Jean de Mau- 
 rienne from the lower valley; and it forms the great 
 feature in the neighbourhood of St Michel and St Martin 
 de la Porte. From the old road by La Traverse its 
 appearance is also particularly striking. It owes this 
 conspicuous character, not only to its great size, but also 
 to its singularly isolated character. It projects, as before 
 observed, for a length of about three-quarters of a mile, 
 from the sides of the mountains on the north of the 
 valley, across which it completely extends, only leaving 
 between the cliff* in which it terminates and the moun- 
 tains opposite, a narrow gorge, or ' etranglement,' just 
 
 1 The subject of this XevKoVerpoi/ might possibly be illustrated, if an 
 accurate description of the promontory of Leucopetra, now Capo dell* 
 Armi, in Calabria, could be obtained. I have not been able to meet 
 with one. 
 
102 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 sufficient for the passage of the Arc. Its height is pro- 
 bably about 500 or 600 feet, the summit being a kind of 
 long irregular plateau, about 200 yards broad. Its southern 
 extremity is a lofty precipice, springing naturally from the 
 river, but now blasted to afford a passage for the road. 
 Its eastern and western faces are likewise abrupt, both 
 being long ranges of absolute precipice, which render the 
 summit of the rock completely inaccessible from either 
 side. On the north it is also inaccessible, except at one 
 point. Here the defile of La Porte perfectly isolates it in 
 its upper part from the sides of the mountains on the 
 north, to which its lower part is attached. Towards this 
 defile the Rock of Baune presents a fourth precipitous side. 
 It is only from the north-eastern corner of the rock, after 
 having gained the highest point of the defile, marked by a 
 cross, and elevated perhaps 300 or 400 feet above the Arc, 
 that it is possible to reach the plateau on the summit of 
 the rock. 
 
 It will easily be seen, from these details, that the Rock 
 of Baune forms a great feature in the country, and is of a 
 particularly inaccessible nature. Indeed its adaptation 
 for a stronghold renders the term 6')(yp6v particularly appli- 
 cable to it. Had it stood in ancient Greece it might have 
 sustained an acropolis. Its claim to the title of a XeuKo- 
 Trerpouf though more liable to be disregarded, is no less 
 well founded ; for its whiteness, if not a striking, is a 
 peculiar and distinctive characteristic of this rock. It is, 
 indeed, only the northern half of its eastern face which 
 is composed of rock either absolutely white, or of a tint so 
 nearly approaching to whiteness that it may be called 
 white ^. In the southern half of the eastern face of the 
 rock, and in the whole of the western face, the colour 
 is grey. The presence of the white portion of the rock is 
 also the more liable to escape notice, in consequence of its 
 being considerably discoloured by the weather, and par- 
 tially concealed by vegetation. It might only be from 
 
 1 Specimens of this rock, submitted to examination, have proved it 
 to bo gypsum. Some of it is perfectly white ; other portions are of a 
 pinkish white. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 103 
 
 certain situations, La Traverse for instance, that the light- 
 ness of its colour would be so remarkable as to strike par- 
 ticularly the attention of an ordinary traveller. 
 
 Yet, though the greater part of the Rock is not white, 
 its whiteness is, nevertheless, one of its peculiarities. The 
 reason is this : all the surrounding mountains are composed 
 of grey rock of a dark hue. The white stratum, which 
 crops out here through the grey, only appears in the Rock 
 of Baune, and in the eminence immediately on the other 
 side of the defile of La Porte. But this eminence is of an 
 unimportant character, and forms no feature in the country. 
 The only one of the great masses of rock in this neigh- 
 bourhood, distinguished from the rest by its whiteness, is 
 the Rock of Baune. Some of the surrounding mountains 
 exhibit enormous surfaces of naked rock. The precipitous 
 mountain rising from the valley directly to the north of St 
 Martin de la Porte, is one pile of crag, of vast height, from 
 base to summit. The bare precipices on the other side of 
 the Arc are scarcely less remarkable. All these rocky 
 masses are of a sombre grey. The Rock of Baune, though 
 also in a great measure of that colour, is distinguished 
 from them in a singular manner by its containing the 
 white rock, which protrudes through the grey^ 
 
 1 The rupture of the upper strata here, and the protrusion of the 
 lower, a convulsion signalized by the formation of the Rock of Baune, 
 is noticed by Prof. A. Sismonda, in his " Meraoria sui terreni stratificati 
 delle Alpi," published in the "Memorie della Reale Accademia dellc 
 Bcienze di Torino" for the year 1841. The author, after having spoken 
 of the calcareous rocks which compose the mountains of the Upper 
 Maurienne nearly as far down as Modane, thus proceeds : (p. 30) 
 
 * Cessando il calcare, viene di nuovo Tarenaria modificata, che mi 
 parvo superiore ad esso, quindi si entra in mezzo a monti coperti delle 
 roccie gia altrove in questo scritto indicate come le rappresentanti 1* Ox- 
 ford clay, ossia 11 terrene antracitoso superiore. Consistono esse in 
 psammiti, in calcare scistoso, arenarie e poddinghe generalmente con 
 tinta bigia intensa, e racchiudono deposit! d'antracite che si scava in 
 varii punti. Le roccie di questo terrene sono rotte sotto S. Michel, o 
 ne esce il calcare, che forma poi il monte collocate al N. N. O. del 
 paese ; la sua posizione, ed i suoi caratteri assicurano che e la continua- 
 zione del potcntissimo banco di Villet. Ho molto detto sulF ordine di 
 Boprapposizione dei differenti terreni fin qui nominati ; tuttavia credcndo 
 
102 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 sufficient for the passage of the Arc. Its height is pro- 
 bably about 500 or 600 feet, the summit being a kind of 
 long irregular plateau, about 200 yards broad. Its southern 
 extremity is a lofty precipice, springing naturally from the 
 river, but now blasted to afford a passage for the road. 
 Its eastern and western faces are likewise abrupt, both 
 being long ranges of absolute precipice, which render the 
 summit of the rock completely inaccessible from either 
 side. On the north it is also inaccessible, except at one 
 point. Here the defile of La Porte perfectly isolates it in 
 its upper part from the sides of the mountains on the 
 north, to which its lower part is attached. Towards this 
 defile the Rock of Baune presents a fourth precipitous side. 
 It is only from the north-eastern comer of the rock, after 
 having gained the highest point of the defile, marked by a 
 cross, and elevated perhaps 300 or 400 feet above the Arc, 
 that it is possible to reach the plateau on the summit of 
 the rock. 
 
 It will easily be seen, from these details, that the Eock 
 of Baune forms a great feature in the country, and is of a 
 particularly inaccessible nature. Indeed its adaptation 
 for a stronghold renders the term o-)(yp6v particularly appli- 
 cable to it. Had it stood in ancient Greece it might have 
 sustained an acropolis. Its claim to the title of a Xcuko- 
 Trerpouf though more liable to be disregarded, is no less 
 well founded; for its whiteness, if not a striking, is a 
 peculiar and distinctive characteristic of this rock. It is, 
 indeed, only the northern half of its eastern face which 
 is composed of rock either absolutely white, or of a tint so 
 nearly approaching to whiteness that it may be called 
 white ^ In the southern half of the eastern face of the 
 rock, and in the whole of the western face, the colour 
 is grey. The presence of the white portion of the rock is 
 also the more liable to escape notice, in consequence of its 
 being considerably discoloured by the weather, and par- 
 tially concealed by vegetation. It might only be from 
 
 1 Specimens of this rock, submitted to examination, have proved it 
 to bo gypsum. Some of it is perfectly white; other portions are of a 
 pinkish white. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 103 
 
 certain situations. La Traverse for instance, that the light- 
 ness of its colour would be so remarkable as to strike par-- 
 ticularly the attention of an ordinary traveller. 
 
 Yet, though the greater part of the Rock is not white, 
 its whiteness is, nevertheless, one of its peculiarities. The 
 reason is this : all the surrounding mountains are composed 
 of grey rock of a dark hue. The white stratum, which 
 crops out here through the grey, only appears in the Rock 
 of Baune, and in the eminence immediately on the other 
 side of the defile of La Porte. But this eminence is of an 
 unimportant character, and forms no feature in the country. 
 The only one of the great masses of rock in this neigh- 
 bourhood, distinguished from the rest by its whiteness, is 
 the Rock of Baune. Some of the surrounding mountains 
 exhibit enormous surfaces of naked rock. The precipitous 
 mountain rising from the valley directly to the north of St 
 Martin de la Porte, is one pile of crag, of vast height, from 
 base to summit. The bare precipices on the other side of 
 the Arc are scarcely less remarkable. All these rocky 
 masses are of a sombre grey. The Eock of Baune, though 
 also in a great measure of that colour, is distinguished 
 from them in a singular manner by its containing the 
 white rock, which protrudes through the grey*. 
 
 1 The rupture of the upper strata here, and the protrusion of the 
 lower, a convulsion signalized by the formation of the Rock of Baune, 
 is noticed by Prof. A. Sismonda, in his " Meraoria sui terreni stratificati 
 delle Alpi," published in the "Memorie della Realo Accademia delle 
 Bcienze di Torino" for the year 1841. The author, after having spoken 
 of the calcareous rocks which compose the mountains of the Upper 
 Maurienne nearly as far down as Modane, thus proceeds : (p. 30) 
 
 * Cessando il calcare, viene di nuovo I'arenaria modificata, che mi 
 parve superiore ad esso, quindi si entra in mezzo a monti coperti delle 
 roccie gia altrove in questo sci itto indicate come le rappresentanti V Ox- 
 ford clay, ossia il terrene antracitoso superiore. Consistono esse in 
 psammiti, in calcare scistoso, arcnarie e poddinghe generalmente con 
 tinta bigia intensa, e racchiudono deposit! d'antracite che si scava in 
 varii punti. Le roccie di questo terrene sono rotte sotto S, Michel, o 
 ne esce il calcare, che forma poi il monte collocate al N. N. O. del 
 paese ; la sua posizione, ed i suoi caratteri assicurano che e la continua- 
 zione del potentissimo banco di Villet. Ho molto detto sulF ordine di 
 Boprapposizione dei diflferenti terreni fin qui nominati ; tuttavia credcndo 
 
lOi 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 105 
 
 The nature of the country in the neighbourhood of the 
 Rock of Baune must now be considered. 
 
 The precipitous mountain to the north of St Martin de 
 la Forte has akeady been noticed. This mountain forms 
 the extremity of a short chain, extending northwards from 
 St Martin to the Mont des Encombres, where it joins 
 the chain dividing the Tarentaise from the Maurienne. 
 This principal chain, after having stretched in length, first 
 E. and then S. E. to the Hoche Chateau Bourreau, sends 
 out a lateral chain in the direction of St Michel. The 
 
 cib cosa utilissima, non posso proseguire senza fermarmivi ogni qual 
 volta pel metodo impiegato in questa descrizione me no viene Toccasione. 
 Ho tuttora ricordato che li psammiti, certi calcari scistosi, le arenarie, e 
 infinc anche certo poddiughe soprastanno al calcare di Villet, il quale d 
 superiore al liasse.' 
 
 M. Albanis de Beaumont, in his * Description des Alpes Grecques et 
 Cottiennes,' notices the closing of the valley of the Arc by the Rock of 
 Baune, and conjectures that Hannibal was attacked here by the AUo- 
 broges and their allies. He speaks, however, only of the modern way 
 by the side of the Arc, and not of the older way through the defile of 
 La Porte. The following is the passage alluded to : 
 
 *A peu de distance de Tentree do cette vallee, (La Combe do Va- 
 loires) commence une chaino de rochcrs calcaircs, dont les faces sent 
 abruptes, et qui compriment de nouveau la vallee de TArc, au point 
 qu*il y a a peine de la place pour le chemin et lo cours de la riviere ; 
 chaquo fois que j'ai traverse ce defile, qui a environ un mille de long, 
 je me suis rappele cc passage de Tite-Live : ad castrum quod erat caput 
 ejus regionisy etc. etc. Dans cette supposition, la ville de S. Jean seroit 
 le castrum dont Annibal ((ju. Tite-Live,) fait mention, et le defile dont 
 je viens de parler, celui oil les Allobroges et leurs allies se mirent en 
 ambuscade pour attaquer rarriere-garde de co general ; les rochers qui 
 bordent ce defile sent, ainsi que j*ai deja observe, en partie calcaires, et 
 en partie schisteux, lis ofFrent memo une grande variete dans ces deux 
 espoces de pierres ; les unos ont des couches tres dures et fort epaisses, 
 d'aiitres tendres, fissiles, et poreuses ; des troisiemes contiennent des 
 paillettes do mica vcrdatre ou gris fence, ainsi que des rognons de petro- 
 silex, leurs couches sent inclinees a Thorizon sous un angle de soixante 
 dcgres; les matieres calcaires qui composent ces rochers sont separees 
 par des couches argilleuses d'uno grande durete, qui se decrepitent en 
 petits feuillets d'un pouce d'epaisseur; celles qui sont schisteuses se 
 montrcnt en grande masse du cote de la rive gauche de I'Arc, elles re- 
 posent sur une roche compacte et tres dure, qui no fait effervescence 
 que tres lentomcnt avec les acides.' (Vol. n. p. 626.) 
 
 three chains, extending in this manner from St Martin de 
 la Porte to St Michel, enclose a large extent of habitable 
 ground, divided into two parts by a small stream, sunk in 
 a gully, and flowing into the Arc through the plain of St 
 Michel, which lies to the east of the Hock of Baune. 
 Another small stream, the source of which is in the Mont 
 des Encombres, flows down from that mountain in a deep 
 impracticable gorge, with precipices on both sides, and 
 forms the impassable western boundary of the habitable 
 ground on the mountains. This gorge terminates imme- 
 diately behind St Martin de la Porte, below which place 
 the stream may be passed without any difficulty. To 
 fticilitate the description of the country, it will be conve- 
 nient to give names to the two divisions of this tract of 
 mountain-ground. The western portion, comprised be- 
 tween the two streams, may be called 'the Hidge of 
 Baune ;' the eastern portion may be designated ' the Slope 
 
 of La Traverse.' 
 
 A path, leading from the valley of the Arc into the 
 Tarentaise, runs along the Ridge of Baune, and crosses 
 the chain which separates the two provinces by the Col 
 des Encombres, a pass which is considerably frequented. 
 The first descent from this pass into the Maurienne is 
 steep, though the pasturages reach nearly to the summit 
 of the Col. The steep descent subsides before reaching 
 Plan Villard, the first village on the road. From thence 
 to Baune the descent is gradual, lying along the back of 
 the ridge, here from about half a mile to a mile broad. 
 At Baune the ridge separates into two branches, divided 
 by a hollow. The first and shortest extends towards St 
 INIartin de la Porte. The descent between these two 
 places is quick, through a series of cultivated fields. ^ The 
 longer branch of the Ridge of Baune terminates in the 
 Rock of Baune, which is cut ofl* from the upper half of 
 this arm of the ridge by the defile of La Porte. On these 
 upper heights there stands by itself a conspicuous chapel ; 
 a circumstance from which the heights may be conve- 
 niently spoken of as * the heights of the chapel.' 
 
 The slope of La Traverse declines at first very rapidly 
 from the crest of the chain on the east ; but, after a long 
 
106 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 descent, the steepness abates, and a plateau or terrace of 
 considerable breadth is presented, extending in length 
 from La Traverse nearly to Villard Piitier. The path 
 between these two places is on a level almost all the way. 
 There is a slight ascent required on leaving Villard Putier, 
 and an equally slight descent before arriving at La Tra- 
 verse. This long plateau between La Traverse and Vil- 
 lard Putier, may be called the * Terrace of La Traverse.' 
 From its v/estern edge, the slope declines steeply again to 
 La Simbran and Villard Bernond. A narrow strip of 
 ground is then left, along which the path ascends from 
 Villard Bernond to Villard Putier. On the west this 
 strip of ground is abruptly cut short by the gully, which 
 forms the channel of the stream already mentioned. 
 
 It is at St Martin de la Porte that the old road was 
 obliged to leave the valley and ascend the mountain-sides. 
 From St Martin it leads along the slopes with a gradual 
 rise for above half a mile, directing its course towards the 
 entrance of the defile of La Porte. After crossing the 
 hollow which divides the two arms of the Eidge of Baune, 
 it runs upon the abrupt side of the Heights of the Chapel 
 for about 200 yards : it then reaches the entrance of the 
 defile, the * Porte ' itself, a passage between the rocks only 
 a yard or two in breadth. The defile within the * Porte ' 
 is wider, though still very narrow. Its length is about 300 
 yards ; the path through it rises all the way ; and the 
 eastern or upper extremity of the defile is nearly as con- 
 tracted as the western. For the 300 yards through the 
 defile, and the previous 200 along the side of the Heights 
 of the Chapel, the path is exposed throughout to be swept 
 by rocks rolled down upon it from the abrupt declivities 
 of those heights. On the side of the Rock of Baune the 
 acclivity is precipitous or very steep, and the defile equally 
 dangerous. On emerging from the defile, the path turns 
 to the left, and passes through Le Village de la Porte 
 along the sides of a slope, eventually crossing the gully, a 
 task of some little difficulty, to Villard Bernond. From St 
 Martin to Villard Bernond, the distance is about a mile 
 and a half. It is about the same distance from Villard 
 Bernond to La Traverse, a great ascent being requisite be- 
 
 MMH 
 
 i n MBia i JM i i 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 107 
 
 tween the two places. For nearly half the distance the 
 mountain-side is very steep, and might be swept from 
 above by rolling rocks. From La Traverse to Le Thyl 
 the way is nearly on a level, and crosses the shoulder 
 of the lateral chain extending from the Eoche Chateau 
 Bourreau towards St Michel. From Le Thyl the path de- 
 scends, having ceased to be flanked by practicable ground 
 after leaving the neighbourhood of La Traverse. Beyond 
 this last village, which is also the termination of the 
 ascent, no enemy could easily collect in any force on 
 the mountain- sides above the road. The perpendicular 
 elevation of La Traverse above the Arc is probably about 
 
 1500 feet. 
 
 Now the features of the country in the neighbour- 
 hood of the ' strong white rock,' as deduced from the 
 narrative of Polybius, are : 
 
 (1) A difficult and precipitous ravine, through which 
 
 the road must pass. 
 
 This is found in the defile of La Porte, a narrow 
 passage to which the expressions -^apahpa, and (pdpay^ 
 SvafiaTos Kai Kprjfxvdh^^ ai'^ perfectly applicable. 
 
 (2) A range of acclivities overlooking the road, and 
 liable to be swept, in one or more places, by rocks set in 
 
 motion from above. 
 
 For about three miles, from St Martin to La Traverse, 
 the road is flanked by acclivities, a considerable part of 
 which, as mentioned just above, might be swept by roll- 
 ing rocks. 
 
 (3) A large extent of practicable ground above these 
 
 acclivities. 
 
 This has been already shewn to be the case in the 
 
 description of the country. 
 
 (4) A position on this ground, at some distance from 
 the road, where 20,000 men could be stationed, so as to 
 protect completely the flank of an army marching along 
 the road beneath, and such as would not expose them- 
 selves to be attacked at any serious disadvantage, by an 
 enemy lying beyond them relatively to the road. 
 
 The position indicated in the Plan would be such a 
 position. A line extending across the back of the Eidge 
 
108 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 of Bauiie, through the village called Le Mollard, would 
 not be exposed to attack at any important disadvantage 
 from the side of Plan Villard. The continuation of this 
 line through Villard Putier, and along the Terrace of La 
 Traverse, would also be exposed to no serious danger of 
 attack from above, if the position were held by a sufficient 
 force. The whole length of this position, from behind La 
 Traverse to the gorge on the west of the Ridge of Baune, 
 would exceed two miles, and might well require 20,000 
 men for its occupation. The whole length of the road 
 from St Martin to La Traverse is, it will be at once seen, 
 completely protected by it. Its distance from the road is 
 also considerable, being, except near La Traverse, more 
 than a mile. 
 
 AVhen Hannibal had broken up his encampment in the 
 neighbourhood of Allevard, he proceeded on his march for 
 three days without experiencing any hindrance, but was 
 attacked on the fourth day at the Hock of Baune. The 
 third night would probably have been spent at St Jean de 
 Maurienne. On the first night after leaving Allevard, he 
 would reach the neighbourhood of Chamoux ; on the 
 second, that of La Chapelle. Now the entrance of the 
 valley of the Arc is at La Croix d'Aiguebclle : at this point 
 Hannibal would, in all probability, leave the country of the 
 Allobroges, and enter that of another tribe ^ Accordingly, 
 we find that it was on the evening of the first day's march, 
 or the morning of the second day's march, that Hannibal 
 met the deputation of the Alpine Gauls ; so that the first 
 night would naturally have been spent upon their frontier, 
 which they might either have crossed to meet Hannibal, or 
 else have awaited his arrival within their borders. As the 
 night, passed by the Carthaginians near Chamoux, was the 
 third after the capture of Allevard, ample time would have 
 been given for the news of the defeat of the Allobroges to 
 have reached the Lower Maurienne, especially if the Intel- 
 
 1 The lower valley of the Arc seems to have been inhabited, in the 
 time of the Romans, by a tribe called Garocoli or Meiulli. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 109 
 
 ligence was brought by fugitives across the mountains, 
 llie approach of the Carthaginians might easily have been 
 known at St Jean de Maurienne, probably the chief town 
 of the tribe, in the middle of the day during which Han- 
 nibal remained at Allevard. On that day the Gauls might 
 have determined on their measures, and resolved to send 
 the deputation to meet Hannibal. The following day 
 Hannibal began his march from Allevard, and the Gallic 
 deputation, probably, left St Jean. On the evening of the 
 same day, the Gauls and Carthaginians perhaps met near 
 Chamoux, about 27 English miles from St Jean de Mau- 
 rienne, and 14 from Allevard ; or else, as seems more pro- 
 bable, the conference may have taken place on the following 
 morning. At this meeting a treaty was made between the 
 two paHies, and Hannibal began to enter into the lower 
 valley of the Arc, the territory of the Gallic tribe whose 
 envoys he had received. 
 
 In two days more Hannibal reached St Jean de Mau- 
 rienne; and, on the morning of the fourth day after 
 leaving Allevard, began his march from St Jean through 
 St JuUen towards the defile of La Porte. A band of Gauls 
 led the way in the van, and other bands followed in the 
 rear, where a large number of men was now collected. 
 The' gathering together of these last bands had made 
 Hannibal extremely suspicious of the intentions of the 
 Gauls, and he therefore placed his heavy infantry in the 
 rear, to repel any attack in that quarter. His apprehen- 
 sions for the safety of the van of his army do not seem to 
 have been great. Perhaps the band of Gauls which led the 
 way was a small one ; and Hannibal would not have been 
 able to see that there were others, as was probably the 
 case, waiting to join them on the heights towards the Col des 
 Encombres. Had all the Carthaginian infantry been m the 
 van and centre, and the baggage and cavalry in the rear, the 
 line of march might have been cut through by an attack 
 from the heights of Baune and La Traverse, the van of the 
 army isolated from the rear, if not overwhelmed, and all 
 the cavalry and baggage probably taken or destroyed. 
 
 The result of the attack was not, however, so disastrous 
 to the Carthaginians. When the Gauls at length fell upon 
 
110 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 the rear, which they probably did between St Julien and St 
 Martin de la Porte, they were able to make no impression 
 on the heavy infantry of Hannibal. The valley is here 
 open, though rather broken and irregular; so that the 
 heavy infantry would fight at no great disadvantage on 
 account of the ground, while their superior weapons and 
 military skill would be sufficient to turn the scale in their 
 favour. The position they took up would perhaps be a 
 little to the east of a small hamlet called Crozat. Here 
 their left might rest on the Arc, and their right upon the 
 side of a mountain rising in sheer precipices, along which 
 it would be impossible for an enemy to move and turn 
 their position. Drawn up thus across the valley, they 
 would resist with success all the efforts of the Gauls in 
 the rear. 
 
 While the Carthaginians thus repulsed the enemy upon 
 their rear, the leading columns, composed of cavalry and 
 baggage-animals, were destroyed by an attack on their 
 flank. How far the Gauls allowed the Carthaginians to 
 proceed along the mountain-road between St Martin and 
 La Traverse is uncertain. All that left St Martin were 
 probably overwhelmed, either crushed beneath rocks rolled 
 down the steep declivities, or killed by great stones, which 
 the Gauls, descending the more gentle slopes, would be 
 enabled to hurl at them from a distance of a few yards. 
 The rest of the horses and other animals would be obliged 
 to remain at St Martin de la Porte, until the passage of the 
 defile, and the road along the mountains, could be made 
 secure. 
 
 The safety of the rear being by this time assured, 
 Hannibal was enabled to execute in person an attack upon 
 the heights near the defile of La Porte. An attack di- 
 rected immediately upon the defile was not likely to have 
 been attended with success, the ' Porte ' being merely a 
 narrow gap in a very steep ridge. The first point to be 
 captured would be the ground about the village of Baune, 
 where the arm of the ridge, which forms the Heights of the 
 Chapel and the Rock of Baune, branches off from the main 
 heights. Though it is a continuous ascent from St Martin 
 to Baune, yet the ground is not difficult, and Hannibal's 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 Ill 
 
 light infantry, being in great force, would eventually suc- 
 ceed, though not without loss, in driving back the Gauls. 
 The capture of Baune would give Hannibal possession of 
 the Rock of Baune and the Heights of the Chapel ; for 
 neither of these positions, lying below Baune, could be 
 defended against a superior force with any chance of success. 
 The Gauls, who had previously held them, would be 
 obliged to retreat in the first instance to Villard Bernond, 
 and then to Villard Putier or La Traverse ; for Villard 
 Bernond lies below Baune, and the gully in front of it 
 would not be a sufficient protection against the Car- 
 thaginians. The Gauls, whom Hannibal had driven from 
 Baune would probably retire towards Le MoUard. 
 
 The capture of Baune would ensure a safe march for 
 the rest of the army from St Martin to the gully. To 
 protect the rest of the line of road, from the gully near 
 Villard Bernond to La Traverse and Le Thyl, the posses- 
 sion of the Terrace of La Traverse is also necessary. The 
 way to carry this terrace from Baune seems evident. It 
 would be necessary to advance along the Ridge of Baune 
 as far as the neighbourhood of Le Mollard, then to cross 
 the gully, not difficult of passage in this place, to Villard 
 Putier, and finally to march from Villard Putier towards 
 La Traverse, thus taking possession of the Terrace. To 
 this movement from Baune, the Gauls could have offered 
 no effectual opposition. They would have been obliged to 
 evacuate Le Mollard, Villard Putier, and La Traverse, as 
 Hannibal approached, and retire towards the higher ground. 
 All the heights commanding the road having been thus 
 gained by the Carthaginians, Hannibal would encamp 
 along the line represented in the Plan, having the Gauls 
 above him, and completely protecting the road beneath 
 from any attack on their part. In this position on the 
 heights, more, as already mentioned, than two miles in 
 length, the half of the Carthaginian army may be supposed 
 to have been stationed. 
 
 By the time the heights had been carried, the evening 
 had probably almost fallen : but Hannibal did not wait for 
 day to withdraw the rest of his army from the valley to the 
 high ground. During the night the Carthaginians passed 
 
112 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 through the defile of La Porte, and by the morning of the 
 next day, but not before, they had all ascended to La 
 Traverse, the first columns having perhaps proceeded further 
 on their way. The safety of the Carthaginians was now 
 assured : the Gauls had no longer any opportunity or incli- 
 nation of offering them further molestation, and retired 
 from the contest. They may, perhaps, seeing their posi- 
 tions lost, have retreated in the night. On the fifth day 
 Hannibal rejoined the rest of his army, and resumed 
 his march ^ 
 
 During the remainder of the march to the summit of 
 the pass, the mountaineers never attacked the Carthagi- 
 nians in any force. Small bands, watching their oppor- 
 tunity, fell at times upon the army at different points in 
 the line, and carried off part of the baggage. AVhere the 
 elephants were found, no attack was ever made, their 
 appearance striking the Gauls with terror and astonish- 
 ment. The march seems here to have lain (as might have 
 been anticipated when Hannibal was approaching towards 
 the summit of the pass) through a country thinly peopled, 
 and inhabited by a wild race of men. There are indeed 
 plausible reasons for conjecturing, that these Gauls were 
 a different tribe from those who dwelt in the Lower Mau- 
 rienne : for they appear to have had no share in the battle 
 of the Eock, nor to have had any previous acquaintance 
 with the Carthaginians, or connexion with the other 
 Gauls 2. 
 
 After the toils of the battle, and the subsequent night- 
 march or watch, the progress of the Carthaginians might 
 be expected to be a little retarded. The evening and 
 night of the sixth day would probably be spent around 
 Modane : the seventh day would bring them to Bramans ; 
 and on the eighth night a large part of the army would 
 
 1 The supposed scene of this contest, and of the operations connected 
 with it, is very well observed from the summit of the rock of Baune. 
 
 2 This conjecture is probably well founded. In the Roman times 
 the Upper Maurienne was inhabited by the Cottian Caturiges, whoso 
 lowest village seems to have been Orelle, situated four miles above La 
 Traverse. See below, Chap. xi. 
 
 The Countrv 
 /// tJic NeioJihonHiood of Hie 
 
 ROCK or BAUNE 
 widitlie Supposed Positions 
 
 CARTKAOIN'IAHS h GAULS 
 
 on tfte evening of' Hie . 
 4*:^ Day's niarcli fi'oiii tlu* Town. 
 
 1 '^ 
 
 2 t4i 
 
 3 ri\ 
 
 6 Fk 
 
 ■diaainian lieatj- Jiuhntrj' 
 
 ihoainian Cdvalr^'.Elephtints: and haq^aije. 
 
 ition of Hannibal on tiie heiohtji with half the Car^iaqinian armv 
 
 4 Oil^uls in -the rallev in the rear of the Carthaqinians 
 
 ilf on tiie heights on the flank of the Carduipinians 
 
 •itions Of' the Gauls '5) whenJLmnihal beqan his march to carrv -die heiahts 
 f%Jioad .,.. r-., Col'^Med 
 
 ■ 
 
 Lit \eofJLjumibals march ah nq the heiahts wiA half his armv Oranqe 
 
 ZkQ .Walker Sc . 
 
112 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 through the defile of La Porte, and by the morning of the 
 next day, but not before, they had all ascended to La 
 Traverse, the first columns having perhaps proceeded further 
 on their way. The safety of the Carthaginians was now 
 assured : the Gauls liad no longer any opportunity or incli- 
 nation of offering them further molestation, and retired 
 from the contest. They may, perhaps, seeing their posi- 
 tions lost, have retreated in the night. On the fifth day 
 Hannibal rejoined the rest of his army, and resumed 
 his march ^ 
 
 During the remainder of the march to the summit of 
 the pass, the mountaineers never attacked the Carthagi- 
 nians in any force. Small bands, watching their oppor- 
 tunity, fell at times upon the army at different points in 
 the line, and carried off part of the baggage. AVhere the 
 elephants were found, no attack was ever made, their 
 appearance striking the Gauls with terror and astonish- 
 ment. The march seems here to have lain (as might have 
 been anticipated when Hannibal was approaching towards 
 the summit of the pass) through a country thinly peoi:)lcd, 
 and inhabited by a wild race of men. There are indeed 
 plausible reasons for conjecturing, that these Gauls were 
 a different tribe from those who dwelt in the Lower Mau- 
 rienne : for they appear to have had no share in the battle 
 of the Hock, nor to have had any previous acquaintance 
 with the Carthaginians, or connexion with the other 
 Gauls 2. 
 
 After the toils of the battle, and the subsequent night- 
 march or watch, the progress of the Carthaginians might 
 be expected to be a little retarded. The evening and 
 night of the sixth day would probably be spent around 
 Modane : the seventh day would bring them to Bramans ; 
 and on the eighth night a large part of the army would 
 
 1 The supposed scene of this contest, and of the operations connected 
 ■with it, is very well observed from the summit of the rock of Baune. 
 
 2 This conjecture is probably well founded. In the Roman times 
 the Upper ISIaurienne was inhabited by the Cottian Caturiges, whoso 
 lowest village seems to have been Orellc, situated four miles above La 
 Traverse. See below, Chap. xi. 
 
 L 
 
 The Couutrv 
 
 t 
 
 /// ////' Xt'inlil'oiiHiOi>ii of tilt' 
 
 ROCK OF BAUNE 
 witliihc Siij)[)()S('(l JV>silions 
 
 I'l' tlir 
 
 CARTHAOINIANS & CAULS 
 ott Hie eveniin] of'thv . 
 4..' Day's iiiai'cl) from \\\v Town. 
 
 •1 Oh 
 
 3 r.K 
 
 •1 da 
 
 (i i'« 
 
 O 
 
 Stuilf i't' oiif Kii)^ Mile 
 
 111 
 
 Hiatiiiiian heaty Jiuaiitrv _.^ „ 
 
 •thaainian Cavalry. IJlephaiil-f. niul inu/ihi'f*^ 
 
 ih'oti ofHanniluiioit the hf/ttftU- uitit htib'ihe lUriiuwinuin ttnn*' 
 
 (//.v /// fhf r,ilh'v in tAerrtuoi'tfie (ortfiaqmumA- 
 
 (Lf <>ii Hie heiqhts on the tUuth oftiie lartiuipiiiUitis 
 
 {ition.v oi' the iiituls '5)whenJhiiuuhalhetfiui his march *» corrv Hie heitjhts 
 
 OHJioaH 
 
 ■••■ ' r— »- 
 
 •e ot'JLiJinihalji nuinh iiL'ntiHie heitihts nith half his amiv 
 
 Col'^JieA 
 Onuuie 
 
 J. &C.VaIk«- St 
 
HANNIBAL'S PASSAGE OF THE ALPS. 113 
 
 most likely have gained the summit of the Little Mont 
 Cenis. On the following morning the rear-guard would 
 arrive, and all the army, with the exception of the strag- 
 glers who afterwards rejoined them, would have accom- 
 plished the ascent of the Alps. The transactions which 
 took place on the summit of the pass, and during the 
 descent, will be considered in the next chapter. 
 
 "\ 
 
 8 
 
 iriirtri1iiiiiiit|ihrimWr%i t 
 
CHAPTER VIII. 
 
 Application of the conditions deduced from Polybius* narrative to the 
 route of the Mont Cenis concluded. — Encampment of Hannibal on 
 the Plateau of the Little Mont Cenis. — Point from which the plains 
 of Italy were most probably seen. — Precipitous character of the 
 descent from the Mont Cenis into Italy. — Probable situation of 
 the place where the path was destroyed. — Arrival at the commence- 
 ment of the plains of Italy. — Siege and capture of Turin. 
 
 THE summit of the Mont Cenis has already been partly 
 described. It consists of two large plateaux, between 
 6000 and 7000 English feet above the sea-level, surrounded 
 by mountains elevated, in some instances, more than 4000 
 feet above the plateaux beneath. The longest of the two 
 plateaux is that of the Great Mont Cenis, extending five 
 miles in length from La Ramasse to La Grande Croix. On 
 the left, towards the east or north-east, this plateau is 
 bounded all the way by the bases of the mountains ; on 
 the right it is bounded, first by the mountains, then by the 
 lake, and finally by a range of hills, running from the lake 
 to the neighbourhood of La Grande Croix, and terminating 
 towards the south in a precipice of great height, which 
 rises from the little plain of St Nicholas. This latter plain 
 lies about 500 feet below La Grande Croix, and is sepa- 
 rated from it by the precipitous descent called the Fourmi- 
 guier. To the W. or S.W. of the lake, extends the plateau 
 of the Little Mont Cenis, about three miles and a half in 
 length, and bounded on both sides by ranges of mountains. 
 The two plateaux form, in a manner, three of the arms of 
 a cross, which meet in the lake. Besides these two pla- 
 teaux, there is a small valley or hollow upon the mountain. 
 It is enclosed between the hills which run from the lake to 
 La Grande Croix, and the mountains on the west. A 
 small stream flows down this valley into the lake. The 
 path from the Col of the Little Mont Cenis, which is at the 
 furthest extremity of the plateau of the same name, leads 
 along the plateau nearly to the lake. It then turns to the 
 
 ; 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 115 
 
 right and follows the little valley behind the hills, which 
 conducts it to a depression in that range of heights, 
 through which it descends upon La Grande Croix, at the 
 extremity of the plateau of the Great Mont Cenis. 
 
 The first condition relating to the summit of the pass 
 is this : 
 
 V. A place suitable for the encampment of Hanni- 
 baPs army must be found at the summit of the pass. 
 
 The plateau of the Little Mont Cenis perfectly satisfies 
 this condition. Its great length and ample width would 
 afford abundant room for the encampment. 
 
 The next condition is the following : 
 
 VI. From a point, probably not far from the road 
 over the summit of the pass, but yet not upon the ground 
 where the Carthaginians would encamp when they first 
 reached the summit, the plains of the Po, and, in all 
 probability, the Apennines also, ought to be visible. 
 
 The ground where the Carthaginians would enccmp, 
 when they first reached the summit of the pass, would 
 be the plateau of the Little Mont Cenis. From thence 
 no part of Italy, nor even any opening in the encircling 
 ranges of mountains, can be perceived. It seems com- 
 pletely shut in by the Alps on every side. On leaving 
 the plateau of the Little Mont Cenis for La Grande Croix, 
 the path turns sharply to the right, and eventually passes 
 over the hills often mentioned, at a point where there 
 is a depression in the chain. Turning to the south, 
 along the crest of the heights, from this point, so as to 
 ascend out of the hollow through which the path runs, and 
 thus arrive upon the long summit of the ridge, the traveller 
 will gain a prospect of Italy in the course of some five 
 minutes. The view is better seen from the southernmost 
 extremity of these eminences, a walk of a few minutes 
 further. The part of the hilly range from whence this 
 prospect is gained, and which lies to the south of the de- 
 pression through which the path runs, forms a ridge about 
 half a mile in length, without any definite head rising 
 above the general level of its summit. It presents a very 
 steep slope towards La Grande Croix, and terminates, 
 
 8—2 
 
116 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 / 
 
 as before mentioned, above the plain of St Nicholas, in a 
 very lofty precipice. 
 
 From the crest of this ridge it may be conjectured 
 that Hannibal pointed out Italy to his army. Great 
 numbers might have collected on the long summit, so that 
 many hundreds of men might have looked upon Italy at 
 once. There is no difficulty whatever in arriving here, 
 and the view is seen from a point so close to the path, that 
 the fact of such a prospect being visible could hardly be 
 unknown to those who were acquainted with the pass. It 
 is indeed only a very small portion of Italy that is de- 
 scried ; but the existence of any extensive prospect does 
 not seem to be required by the narrative. In fact, if we 
 suppose the action, intimated by the word euSeiKvvfxevos, to 
 have been a natural, and not merely an oratorical, gesture, 
 we should be led to imagine that only a small part of the 
 plains was visible ; for to any very large expanse it would 
 have been superfluous to direct attention. Besides, any 
 prospect of Italy, however limited, would have been suffi- 
 cient for Hannibal's purpose. It would have proved to the 
 Carthaginians, by visible demonstration, that their extri- 
 cation from the Alps was at hand, that the mountains were 
 about to terminate, and that the plains of Italy were 
 almost gained. 
 
 It is towards the S. E. that the view is open. The 
 country seen would thus be the district to the east of the 
 Po, and the south of the Tanaro, where the cities of Alba 
 and Acqui are situated. This part of the plains is inter- 
 sected by several ranges of hills, one of which may be dis- 
 cerned from the point of view on the Mont Cenis, even 
 after the hazes, so prevalent in the plains of the Po during 
 a great part of the day, especially in summer, have effiiced 
 the prospect of the flat country. In the extreme distance 
 the chain of the Apennines closes the view, and would have 
 offered to Hannibal, as previously explained, the means 
 of indicating the position of Rome'. 
 
 1 Another point on the Mont Cenis, from which the plains of Italy 
 are visible, is mentioned by M. Grosley, who wrote in the middle of the 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 117 
 
 On the morning of the ninth day the whole Cartha- 
 ginian army had gained the summit of the Little Mont 
 Cenis. During the remainder of that day, and during the 
 following night, they remained encamped on the plateau of 
 the Little Mont Cenis. On the morning of the tenth day 
 Hannibal found them in a state of great despondency. 
 One resource only, but that an effectual one, and such as 
 could never have been anticipated by the Carthaginians, 
 offered itself to Hannibal, as a means by which he could 
 restore the courage and fortitude of his army. He had 
 become aware, probably by the report of the Cisalpine 
 Gauls who then accompanied him, and who had previously 
 crossed the pass to meet him on the Rhone, that there was 
 a point on the mountain, by the side of the path, some two 
 or three miles from the plateau of the Little Mont Cenis, 
 from which the plains of Italy could be discerned. To this 
 point, accordingly, Hannibal led his army, or a part of it, 
 and pointed out to them the plains of the Po, and the 
 range of the Apennines, behind which, as he stated, lay 
 the city of Rome itself. He dwelt also upon the friendly 
 disposition towards them of the inhabitants of the country 
 then lying beneath their feet, and succeeded eventually, 
 to a certain extent, in dispelling the gloom which had 
 fallen upon their spirits. By this movement to obtain the 
 prospect of Italy, the position of the Carthaginian encamp- 
 ment on the tenth day would be a little altered from what 
 
 last century on the subject of Hannibars route across the Alps. The 
 passage is given by Ukert in his Geographie, il. 2. p. 600. 
 
 L'espece de coupe que forme le plateau du Mont Cenis, est bordee 
 de falaises tres elevees, et ainsi il n occupe pas, au pied de la lettre, le 
 Bommet de la montagne. C'est a mi-c6te d'une de ces falaises, a la 
 hauteur du Prieure, qu'on decouvre les plaines du Piemont, et c'est do 
 la qu'Annibal put les montrer a son armee. 
 
 I believe this point of view to be a remarkable kind of shelf on the 
 eastern side of the mountain which lies S. W. from the lake. It is 
 easily recognized from the high road over the Mont Cenis. This shelf 
 is distinctly seen, by the aid of a telescope, from La Loggia, a village 
 about seven miles to the south of Turin. The path from the Little 
 Mont Cenis runs at the base of the mountain, on the side of which it 
 is situated. The ascent to the shelf is steep, and in time of snow almost, 
 if not quite, impracticable. 
 
> 
 
 118 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 119 
 
 it was on the ninth : for, from the summit of the hills 
 from which Italy was seen, it is not likely that the Cartha- 
 ginians would have been led back to the plateau of the 
 Little Mont Cenis, but would rather have descended to 
 La Grande Croix. On the tenth night the army probably 
 lay encamped on the eastern extremity of the plateau of 
 the Little Mont Cenis, around La Grande Croix at the 
 southern extremity of the plateau of the Great Mont Cenis, 
 and on the ground between these two positions. On the 
 eleventh morning they began, from La Grande Croix, their 
 descent into Italy, and left the summit of the Mont Cenis, 
 on which they had spent two days and nights. 
 
 The accordance of the paSs of the Mont Cenis with the 
 seventh condition, that ' The commencement of the plains 
 of Italy must be less than three days' march from the sum- 
 mit of the pass,' has already been shewn in an earlier 
 chapter (p. 72). We pass on therefore to the eighth con- 
 dition. 
 
 VIII. The first part of the descent from the pass on 
 the Italian side must be of a precipitous character. 
 
 The precipitous character of the descent into Italy by 
 the old road over the Mont Cenis, has been already men- 
 tioned (p. 87). It is one of the most remarkable features 
 of the pass, and is particularly noticed in many descrip- 
 tions of it^ Between La Grande Croix, where the descent 
 
 1 Enfin, a rextremite de la plaine est Tauberge de la Grande-Croix, 
 ou les Toyageurs s'arretent souyent pour se reposer ou se rechauffer. 
 C'est de \h que commence la descento du cote du Piemont. La pente 
 de Fancien chemin qui passait par Ferrieres, et aboutissait k la Novalaisey 
 ou Ton faisait remonter les voitures, etait si rapide, quo plusieurs voya- 
 geurs ont fait la peinture la plus effrayante des dangers qu'elle presen- 
 tait.—Richard, Guide en Savoie et en Piemont. 
 
 Dalla Novalesa ultimo villaggio del Piemonte, tale era I'antica strada ; 
 per una rapida e sparentosa salita, a svolte serpeggianti si giungera al 
 villaggio delle Ferrieres, d'onde oltrepassato il piccolo piano di S. Niccolo 
 ove ancora si redono i limiti che segnavano i confini della Savoia e del 
 Piemonte, montate le erte rampe a zig-zach dette il Fourmiguier, si veniva 
 alia Gran Croce, povero e scarso abituro, e traversatono il piano si met- 
 teva capo alia Ramassa, d'ondo scorrendo pel dorso del monte si calava 
 a Lanslebourg.— /ii»2erano Postale degli Stati Sardi, 
 
 begins, and the village of La Novalese, near which it ter- 
 minates, the road falls, in the course of six miles, more 
 than 3600 feet; so that the rapidity of the descent must 
 necessarily be considerable and long-continued ^ 
 
 We now come to the ninth condition. 
 
 IX. At a place, within half a day's march from the 
 summit of the pass, on the Italian side, the path must lie 
 along a precipitous mountain-side, liable to be broken up 
 for a length of nearly 300 yards (English) ; and the country 
 in the vicinity should be of a nature adapted to Polybius' 
 account of the events which took place in the neighbour- 
 hood of the broken path. 
 
 A remarkable passage in a geographical work of the 
 last century seems to indicate clearly the situation of this 
 place, where Hannibal's path had been destroyed by a 
 landslip 2. The passage in question runs thus^; 
 
 * From the inn, called La Grand' Croix, on account of 
 the wooden cross which stands by its side, and which forms 
 the boundary between Savoy and Piedmont, the descent 
 begins. On descending, there is found a plain, enclosed 
 
 M. Deluc also (Histoire du Passage des Alpes par Annibal, p. 280,) 
 notices the precipitous character of the descent from the Mont Cenis, a 
 circumstance which he brings forward as an argument against the suppo- 
 sition of HannibaFs having crossed this pass. See below. Chap. X. 
 
 1 Height of La Grande Croix above the sea, 6211 English feet; of 
 La Novalese, 2577. 
 
 2 Biisching's Geography of Italy, enlarged and improved edition. 
 Vemce, 1780. Vol. i. p. 78. 
 
 3 Dall osteria, detta la Grand^ Croix, dalla croce di legno, che v*e 
 accanto, e che forma il confine tra la Savoja ed il Piemonte, incomincia 
 la scesa. Nello scendere ritrovasi una pianura rinchiusa fra* monti, detta 
 la Plaine de S. Nicola, passata la quale v*e una scesa, che tempo fa i 
 macigni, ed i sassi talmente rendevano impraticabile, che colore, i quali 
 solevano portar in sedia i viandanti, eran costretti a scender da* sassi a 
 salti, come da tanti gradini. In questa discesa inevitabile, presso il No- 
 valese nel Piemontese v'erano 3 o 4 luoghi, ove 11 sentiero fiancheggiato 
 da precipizj altissimi era strettissimo, e Tacque talmente aveano smosso 
 il terrene, che la sedia colla persona portatavi, per la meta pendeva in 
 aria sul precipizio. Ma sotto il regno di Emmanuele IIL fu fatta una 
 strada nuova, ove non c*e pericolo alcuno ; nondimeno v*e uno spazio di 
 circa 16 miglia, ove il passaggiero e obbligato a farsi portare in sedia. 
 
 \ 
 
/ 
 
 120 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 by the mountains, called La Plaine de S. Nicola; this plain 
 being passed, there is a descent, which, at one time, the 
 stones and rocks rendered impracticable to such a degree, 
 that those who were accustomed to transport traveUers in 
 chairs, were obliged to descend by leaping from rock 
 to rock, as it were down so many steps. Upon this inevit- 
 able descent, towards La Novalese, there were three or four 
 places, where the path, flanked by very lofty precipices, was 
 exceedingly narrow, and the waters had broken away the 
 ground to such an extent, that the chair, with the person 
 carried in it, hung half in air over the precipice. But under 
 the reign of Emmanuel III. (1730—1773), a new road was 
 made, upon which there is no longer any danger ; never- 
 theless there is a distance of about sixteen miles where the 
 traveller is obliged to be carried in a chair/ 
 
 This part of the old road is again referred to in the 
 same work. It is said, in the description of La Ferriere, 
 that the path leading down to La Novalese was sometimes 
 contracted by the precipices to a width of no more than 
 one foot\ 
 
 Here then we have direct evidence that, between La 
 Ferriere and La Novalese, the path over the Mont Cenis, a 
 century ago, lay along a precipitous mountain-side liable 
 to be broken up by the falling away of the ground. In the 
 English translation of Busching (London, 1762) it is men- 
 tioned that these three or four dangerous places were onlv 
 a fQyy paces in length : they would no doubt be the worsl: 
 parts of the crumbling precipice, where it was not possible 
 to construct any solid road. 
 
 The mule-road of Emmanuel HI., a broad and strongly- 
 constructed way, runs from La Ferriere to La Novalese on 
 the right bank of the Cenise : the old road, abandoned on 
 account of the dangers caused by the landslips, ran on the 
 left bank of the stream. This last path is still partially 
 
 1 Ferriere, Ferrera, o Ferrara, borgo che giace alia met^ della strada 
 fra la Grande Croix, e la cMik di NoFaloso, cioe nelle falde del Monto 
 feenis. La strada fra questo luogo, e fra Novalese in piCi luoghi e molto 
 nstretta fra' dirupi, e passa talvolta in mezzo a' precipizj, che non las- 
 ciano piCi d'un piede di larghezza. ^ i J» 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 121 
 
 # 
 
 used by the inhabitants of the country. It is a distance of 
 about three miles from La Ferriere to La Novalese. For 
 the first mile on the road there was no difficulty, or none 
 of any importance : but with the second mile the dangers 
 began. The Cenise had previously flowed through a series 
 of fields, at no great distance below the path. It now sank 
 in a narrow gorge, and left the path high above it, running 
 along the face of an abrupt declivity composed of earth 
 intermingled with rocks, and continually liable to crumble 
 away. Here the old road, still remaining, forcibly recalls 
 to mind the narrative of Polybius, where he relates how 
 Hannibal found his path destroyed by a landslip, and was 
 eventually obliged to reconstruct it, by building up the pre- 
 cipice. For the old road is here, in many places, supported 
 on terraces, such as Hannibal must have raised along the 
 mountain-side^. These have, however, fallen away in many 
 places, the path being now only partially used by the inha- 
 bitants, and having been long abandoned by travellers. 
 From the point where it first begins to skirt the precipice, 
 about a mile below La Ferriere, it runs nearly in a straight 
 line, gradually descending, for about three hundred yards. 
 It then meets with a gully, laying open the mountain-side, 
 and quite impassable where the path first encounters it, on 
 account of the sheerly precipitous character of its eastern 
 face. Repelled at this point, the path makes an angle, 
 and returns for above a hundred yards along the precipice 
 towards La Ferriere, thus gaining a lower level. It then 
 resumes its original direction, and, after a course of about 
 three hundred yards more, reaches the gully again, and 
 succeeds in crossing it. At the point where the gully is 
 crossed, and for a considerable distance beyond, the path, 
 never used now, is quite destroyed. Nothing but a mere 
 cattle track is left, though fragments of the old terrace- 
 walls may still be perceived where the path has been 
 destroyed. Proceeding towards La Novalese, the moun- 
 tain-side, though steep, becomes less precipitous, and the 
 path is well preserved. Eather more than half a mile 
 beyond the gully, the head of the valley of La Novalese is 
 
 1 See above, pp. 55 — 57. 
 
 ■gJM 
 
122 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 reached, and another mile across the fields brings the 
 traveller to the village itself^ 
 
 It would be on the first part of this descent that Han- 
 nibaPs path was broken up. The length of the first arm 
 of the ancient zigzag road, about 300 yards, seems to 
 identify this part of the mountain-side with the crumbling 
 precipice, recently broken up, which Polybius mentions as 
 nearly a stadium and a half across. The whole of the 
 ground on the west of the gully was, in all probability 
 more or less broken up by the landslip which had taken 
 place 2. The distance from La Grande Croix to the com- 
 mencement of this precipice is about four English miles 
 It was, therefore, within half a day's march from the sum- 
 mit of the pass, the hamlet of La Grande Croix, at the 
 southern extremity of the plateau of the Mont Cenis, beino- 
 the point from which Hannibal commenced his descent* 
 The first part of Condition IX. is thus fully satisfied : the 
 latter part, relating to the events which took place in this 
 neighbourhood, now requires consideration. 
 
 These events have been already discussed (pp. 57—60) 
 When Hannibal first found that the path was destroyed, he 
 attempted to continue his march by taking a circuit. He 
 arrived however eventually, on making this attempt, upon 
 a very steep slope of old snow, thinly covered with snow 
 which had lately fallen. It has been conjectured above, 
 from this circumstance, and also because such a line 
 of descent would be natural, that Hannibal attempted to 
 bring his army down by some gully; for in such hollows 
 the old snow frequently lingers, when the ground in the 
 neighbourhood is quite clear. These gullies also form the 
 channels of avalanches, by which, on extraordinary occa- 
 sions, old snow is sometimes found at a very low level, even 
 in the heat of summer. Such a circumstance may be 
 
 1 The whole course of this ancient path is well seen from the later 
 mule-road. The old lino of way may still be followed throughout on 
 foot, though with some little difficulty where the path has been destroyed 
 
 2 The cause of this landslip was, perhaps, the series of heavy rains, 
 
 ^which, as may be gathered from Livy's account of the swellin- of the 
 
 Druentia,' had prevailed in the Alps about three weeks before Hannibal 
 arrived at the broken path. 
 
 / 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 323 
 
 alluded to by Polybius in the words iStov kuI TraptjWayfxevov 
 (Chap. 55), though it is not absolutely necessary to make 
 such a supposition. For, although no permanent snow is 
 now found anywhere near the path between La Ferriere 
 and La Novalese, or indeed upon any part of the Mont 
 Cenis road ; yet, in ancient times, the climate of Europe 
 must have been much severer than it is at present, and 
 permanent snow would have been found on the passes of 
 the Alps at a considerably lower level than where it is now 
 met with. Indeed, we should be led to conclude, from 
 what Polybius says, not only that the summits of the passes 
 were continually covered with snow, but that the fields of 
 snow even reached half-way down the declivities, on both 
 sides of the Alpine passes, towards the level country ; so 
 that not only patches of snow might be found at a con- 
 siderable distance below the summits of the passes, but 
 even the great snow-field itself may have extended to a 
 distance of several thousand feet below the present line of 
 perpetual snow\ It seems, however, necessary to under- 
 stand these statements of Polybius with some modification, 
 and to suppose that he speaks, not of fields of snow, but 
 rather of isolated patches, which are found, at the present 
 day, in many instances, two or three thousand feet below 
 the snow-line, and, in some places, even more 2. 
 
 It may thus be supposed, without violating probability, 
 that, at the time of Hannibal's descent, the gully on the 
 precipitous way between La Ferriere and La Novalese was 
 filled with old snow, compact and hard; and that fresh 
 snow may have lain upon it to the depth of a few inches. 
 Now, if all the ground to the west of the gully was broken 
 up, but the path, as appears to have been the case, re- 
 mained uninjured on the east of the gully ; then the way 
 by which Hannibal would naturally have attempted to 
 
 T<op yap AXttccov to, fiev oKpa koi tu npos ras vrrcp^oXas avi^Kovra 
 TfXccoff abevdpa koi "^ika iravr icrri bia to crvpe^^as inififvfip rrjv xiova Koi 
 Bepovs Kai xfni&voSi ra 8* vtto fxea-rjp t^p jrapapeiap i^ dp,(f)o7p toIp fxepolv 
 v\o(l)6pa Kot d€pbpo(P6pa koi to okop oiKija-tfia ia-Tip. 
 
 2 On the subject of the greater severity of the climate of Europe in 
 ancient times, see Gibbon's Roman Empire, Chap. IX., and Arnold's His- 
 tory/ of Rome, Vol. in. p. 89, and Note M, p. 485. See also below, Chap. X. 
 
124. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 rejoin the path where it was undestroyed, would have been 
 down the gully itself. From the point where the path 
 from La Ferriere first begins to run along the precipice, it 
 would not be very difficult, (for men alone it would be 
 easy) to pursue a directly straight course (represented by 
 the broken line in the sketch) as far as the gully. A great 
 part of the ground which would be traversed is now 
 covered with grass, and quite unlike the crumbling face of 
 the precipice beneath. Hannibal's army however could 
 not cross the gully, for its eastern face is quite precipitous, 
 but would be obliged to attempt to descend by its channel. 
 The calamitous results which Polybius mentions would 
 then naturally follow. For the fresh snow, which, by its 
 yielding character, seemed to give a firm hold to the foot 
 which penetrated it, was only a few inches in depth, and 
 had not adhered to the hard snow beneath. This hard 
 snow the feet of the men did not penetrate ; they slipped 
 and fell, attempted in vain to save themselves by clinging 
 to some support, probably to their companions, rolled 
 down the snow with whatever they had clung to, and were 
 lost down the precipices. The animals, however, heavily 
 laden, struck their feet into the old snow in their efforts to 
 save themselves when they slipped, and remained fixed 
 without the power of extricating themselves, being held 
 fast by the tenacity of the hard-compacted snow. Han- 
 nibal, seeing from these misfortunes that it would be 
 impossible to effect his descent, excepting by the line of 
 the old road which had been destroyed, abandoned his 
 attempt to descend by the gully, and determined to re- 
 construct the broken path. 
 
 When Hannibal had thus relinquished his first project, 
 ecTTpaToweSevcre irepl t^v paxiv, ^lafxrjad/ueuo^ Trjv eir avrPj 
 
 Xt^i^ay he encamped about the back (or crest) of the ridge, 
 having cleared away the snow from it^ Now the descent 
 from the plain of St Nicholas to the head of the valley of 
 La Novalese is effected down a steep hollow, sloping to the 
 east, and bounded on the north and south by very lofty 
 mountains. This hollow, however, is not single, but is 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 125 
 
 divided into two minor parallel hollows by a ridge extend- 
 ing from west to east, and originating a little below the 
 plain of St Nicholas. In the northern of these hollows 
 flows the Cenise ; in the southern, and at a distance of about 
 half a mile from the Cenise, runs a small rill of water, which 
 joins the Cenise near La Novalese. It seems probable that 
 it is this ridge, which is alluded to by Polybius under the 
 name of the pdxi^' Its crest or back, at its eastern ex- 
 tremity, is a nearly level tract of ground, raised on very 
 high precipices above the valley of La Novalese, and having 
 on one side the deep gorge of the Cenise, where the path 
 was destroyed, and on the other a very steep gully, 
 through which the rill of water, previously mentioned, falls 
 down into the valley beneath \ This tract of ground is 
 now almost entirely covered with great blocks of rock, the 
 results, apparently, of some mountain-fall. These heaps 
 would at present disqualify it from being the site of an en- 
 campment ; but their fall may have been subsequent to the 
 time of Hannibal. The fields through which the Cenise 
 flows, between La Ferriere and the commencement of the 
 broken precipice, would have afforded in themselves a con- 
 siderable space of ground for the encampment. Yet the 
 whole army could hardly have found sufficient room here, 
 so that Hannibal would have been obliged to pitch part of 
 his camp on the ridge to the south of these fields. If there 
 were no heaps of fallen rocks at that time on the ground 
 at the eastern extremity of the ridge, it would be there 
 that the most eligible position would have been found for a 
 part of the camp. 
 
 In one day, the path which Hannibal constructed along 
 the precipice, by means of terraces, was sufficiently wide 
 and strong for the passage of the horses and baggage-ani- 
 mals. They were immediately brought down from the 
 encampment above, another encampment was made below, 
 on ground perfectly free from snow, and the animals were 
 then left to find pasturage for themselves in the country 
 round. This second encampment would be made near La 
 
 1 See above, p. 59. 
 
 ^ When Hannibal attempted to descend by taking a circuit, he may 
 possibly have tried to bring his army down by this gully. The gully, 
 however, on the broken precipice itself, seems the more likely place. 
 
126 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 Novalese, immediately at the foot of the precipitous and 
 broken descent ; and in the broad fields at the head of this 
 open valley, the animals would be let loose, and left to seek 
 the fodder, of which they were so greatly in need. The 
 upper and lower encampments would probably be less than 
 two miles apart ; and that near La Novalese would be in 
 sight of the one above. 
 
 On the evening of the thirteenth day, the road was 
 completed for the passage of the elephants, who were by 
 this time almost starved ». They were accordingly brought 
 down, and the whole army was again collected together. 
 In two days more the Carthaginians reached Avigliana, at 
 the commencement of the plains of Italy. The tenth knd 
 last condition relates to this point in the route, where the 
 Alpine march was at last brought to a conclusion. 
 
 X. The plains, into which the road over the pass 
 enters, when it emerges from the mountains on the side of 
 Italy, must anciently have been inhabited by the Taurini. 
 
 Avigliana is generally identified with the ^ Ad fines ' of 
 the Eoman Itineraries, the boundary between the district 
 of the Taurini and the Cottian tribe of the Segusini 
 Hannibal would therefore be, at Avigliana, on the frontier 
 of the Taurini, and at the commencement of the plains 
 which they occupied. 
 
 ^ No part of HannibaPs passage of the Alps was attended 
 with such disasters as the descent into Italy. Although 
 the Carthaginians encountered no hostility on the part of 
 the inhabitants, (for the furtive aggressions mentioned by 
 Polybius seem to have been too insignificant to merit such 
 a name,) yet their losses, in consequence of the snow and 
 the precipitous character of the descent through the unin- 
 habited country, were nearly as great as they were upon 
 the ascent from the banks of the Isere to the summit of 
 the pass. Between La Grande Croix and La Novalese a 
 distance of only six miles, nearly 10,000 men, or one-fourth 
 of the whole army which encamped on the summit of the 
 
 1 It should seem that fodder might have been brought up to the 
 elephants from below. Nothing of the kind however is mentioned by 
 Polybius. ^ 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 127 
 
 mountain, must have perished without meeting a living 
 enemy. Nothing but the extremely precipitous character 
 of the descent from the Mont Cenis can explain this enor- 
 mous loss. Whole hosts, indeed, must have been destroyed 
 by the precipices of the Fourmiguier, between La Grande 
 Croix and the plain of St Nicholas. All this abrupt decli- 
 vity, probably some 500 feet in perpendicular elevation, 
 would have been, at the time of Hannibal's passage, com- 
 pletely covered by the snow, which had lately fallen, and 
 beneath which the path was lost. A vast and trackless 
 precipice of snow, which there was no possibility of avoid- 
 ing, presented thus the only way by which the army 
 could descend. On this slippery declivity, nothing that 
 fell escaped destruction. Multitudes must have been 
 carried down the precipice, before the survivors succeeded 
 in reaching the plain of St Nicholas. On the descent 
 from this plain, similar losses, but probably on a much 
 smaller scale, would have been experienced. The army, 
 however, it is said, bore all these disasters with fortitude 
 because they had already encountered calamities of a like 
 nature. Perhaps, on the steep ascent to the Col of the 
 Little Mont Cenis, the path may have been partially con- 
 cealed by the snow, and some losses have ensued in con- 
 sequence. It was, however, the fresh snow which seems 
 to have fallen while they remained for two days on the 
 summit of the pass, by which the dangers of the descent 
 were so terribly increased. 
 
 At length the progress of the Carthaginians was com- 
 pletely arrested. The path by which they hoped to have 
 descended had been destroyed by a landslip, and they 
 found themselves condemned, in their reduced and starv- 
 ing condition, to fix their encampment in the midst of the 
 snow. Their despair at this calamity, and the bold, and, 
 as it might seem at first, the even rash attempt of Hannibal 
 to bring the army down by a new and untrodden route, can 
 readily be understood. They looked down on the broad 
 fields of La Novalese immediately below them. No snow 
 was there to be seen, and no danger or difficulty was after- 
 wards to be encountered. Could they but succeed in 
 effecting this last descent, all their calamities and priva- 
 tions would be at once greatly mitigated, if not entirely 
 
128 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 terminated. To gain this end, it was well to risk the 
 dangers of an uncertain descent, before embracing the only 
 alternative which remained, of halting, amidst the snow, tiU 
 the broken path could be reconstructed. 
 
 This latter course, however, was found inevitable ; and 
 It was only on the thirteenth evening that the elephants de- 
 scended to La Novalese, where all the army was again 
 collected together. No events of importance characterized 
 the rest of the march to the plains of Italy. Indeed no 
 effectual opposition could have been offered to Hannibal 
 here, on account of the nature of the country. For the 
 valley of the Dora below Susa, is perfectly open : there 
 are no defiles which might be defended, or which would 
 offer opportunities for any serious attack. On the four- 
 teenth night, Hannibal would probably reach the neieh 
 bourhood of the village of S. Giorio, and on the fifteenth 
 all the army would encamp, at the foot of the Alps, about 
 Avighana. Here Hannibal remained for some time, durin- 
 which his exhausted forces recovered their strength and 
 condition^ He then advanced into the plains and laid 
 siege to Turin, (16 Roman miles from Avigliana') the chief 
 town of the Taurini, who had rejected his overtures of 
 amity and alliance. After three days' siege, the town was 
 Uken and all that offered opposition put to the sword. 
 This example of severity brought all the neighbourinjr 
 people to Hannibal's feet. The other Gallic tribes, also, 
 who inhabited the plains, shewed themselves favourably 
 inclined to the Carthaginian alliance ; and Hannibal pre- 
 pared to advance, and strike some great blow, which 
 might fix them permanently to his side, and induce them 
 to join with him in attempting the overthrow of Rome. 
 
 1 In the Roman Itineraries, one measurement (that generally adopt- 
 ed) gives 24 m,les as the distance from Segusio to Ad FiL, and I JL 
 as the distance from Ad Fine, to Augusta Taurinorum. Another mea! 
 snrement g.ves 33 (a plain mistake for 23) miles in the first instance, and 
 
 *h« L? T \ "Z^'S"'""' «'»»''« ^ecurately on the site of Ad fJ, 
 the latter (corrected) measurement would perhaps be nearest the truth 
 By the roads of 1839, according to the Sardinian Itinerary published in 
 ha year, the distance from Turin to Avigliana is ^, Piedmontese miles 
 (16j Roman miles); and the distance from Avigliana to Susa is 12 
 Piedmontese miles (20| Roman miles.) 
 
 / 
 
128 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 terminated. To gain tliis encl, it was well to risk the 
 clangers of an uncertain descent, before embracing the only 
 alternative which remained, of halting amidst the snow, till 
 the broken path could be reconstructed. 
 
 This latter course, however, was found inevitable ; and 
 It was only on the thirteenth evening that the elephants de- 
 scended to La Novalese, where all the army was again 
 collected together. No events of importance characterized 
 the rest of the march to the plains of Italy. Indeed no 
 effectual opposition could have been offered to Hannibal 
 here, on account of the nature of the country. For the 
 valley of the Dora, below Susa, is perfectly open • there 
 are no defiles which might be defended, or which would 
 offer opportunities for any serious attack. On the four 
 teenth night, Hannibal would probably reach the neio-h 
 bourhood of the village of S. Giorio, and on the fifteenU, 
 all the army would encamp, at the foot of the Alps, about 
 Avighana. Here Hannibal remained for some time, durin- 
 which his exhausted forces recovered their strength and 
 conditiom He then advanced into the plains and laid 
 siege to Tunn, (IG Roman miles from Avigliana') the chief 
 town of the Taurini, who had rejected his overtures of 
 anruty and alliance. After three days' siege, the town was 
 taken and all that offered opposition put to the sword 
 Ihis example of severity brought all the neighbourinjr 
 people to Hannibal's feet. The other Gallic tribes, also 
 vvho inhabited the plains, shewed themselves favourably 
 inclined to the Carthaginian alliance ; and Hannibal pre- 
 pared to advance, and strike some great blow which 
 might fix them permanently to his side, and induce them 
 to join with him in attempting the overthrow of Rome 
 
 .A\ • ol r"" ^"'"=f '"'• °"° '""'^•"•cment (that generally adopt- 
 ed) gives 24 miles as the distance from Segusio to Ad Fines, and 1« miles 
 as the distance from Ad Fines to Au<;usta Taurinorum. Another mea! 
 surement g.ves 33 (a plain mistake for 23) miles in the first instance, and 
 
 th^K t 7 \ i^/^'e"""* ^'-n«'« accurately on the site of Ad Fines, 
 the latter (corrected) measurement would perhaps be nearest the truth 
 By the roads of 1839, according to the Sardinian Itinerary published in 
 ha year, the distance from Turin to Avigliana is 9f Piedmontese miles 
 (16 1 Roman miles); and the distance from Avigliana to Susa is 12 
 Piedmontese miles (20^ Roman miles.) 
 
 / 
 
CHAPTER IX. 
 
 Uvy's narrative of Hannibars Passage of the Alps. — Its character. Use 
 
 to be made of it.— Hannibars line of approach to the Alps, as in- 
 dicated by Livy. — His account of the events which took place on 
 the summit of the pass, — The broken path. — Singular story con- 
 nected with it. — Arrival of Hannibal in the country of the Taurini. 
 
 IT will not be necessary to examine Livy's account of 
 HannibaPs passage of the Alps at the same length as 
 that of Polybius, the greater importance and accuracy of the 
 Greek narrative being sufficiently evident. Indeed, Livy 
 had none of the peculiar advantages of Polybius. He 
 flourished about a century and a half after the earlier 
 historian, and had therefore no communication with the 
 contemporaries of the expedition. Neither had he ex- 
 amined the scene of Hannibal's passage, as Polybius had, 
 but appears ignorant, not only of that particular pass, but 
 also of the general nature of the Alpine country. The 
 authority of Livy's narrative is also much weakened by his 
 extremely careless mode of composition. He seems to 
 have collected his materials from various sources, and then 
 to have reported them almost in their crude state, without 
 examining their bearing on one another, or digesting them 
 into a consistent story. He also relates some events of a 
 strange and improbable nature, and is glaringly incorrect 
 with respect to dates. Carelessness and love of the mar- 
 vellous are the chief causes of his errors ; his aim is to 
 produce effect, rather than to ensure accuracy and truth. 
 
 Notwithstanding, however, Livy's faults, his narrative 
 cannot be totally disregarded, or set aside as entirely false. 
 Some true materials must be worked up in it. Indeed, 
 the chief materials are clearly furnished by Polybius, though 
 some additional and independent matter is introduced. 
 A part, at least, of this additional matter, is probably 
 authentic, however disfigured and alloyed by carelessness 
 and extravagance. But, although whatever is clearly ex- 
 travagant must be either rejected or modified, and what- 
 ever contradicts Polybius must be set aside, a considerable 
 
 9 
 
ISO 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 part of the history will still be left, and cannot well be 
 neglected. A body of new facts will thus be obtained. 
 Yet, even with respect to these facts, further precautions 
 may be necessary; for, in the case of such a writer as 
 Livy, the arrangement of his materials, even when the facts 
 they contain may be true, is very likely to be faulty, and 
 may demand re-adjustment. When the materials, on which 
 a history is founded, are collected from various quarters, 
 and arranged by a careless historian, it is not unreasonable 
 to imagine that they may sometimes be wrongly disposed, 
 and that the order in which events actually happened may 
 thus be inverted. The first passage in Livy which will 
 require consideration seems to be vitiated by an error of 
 this nature, and is inexplicable as it stands. In examining 
 his account, it will be necessary partly to undo what he has 
 done ; to resolve his ill-digested narrative into its com- 
 ponent parts, and to arrive in this manner, through the 
 history, at a knowledge of the materials from which it has 
 been made up. 
 
 There seems to be nothing that calls for notice in 
 Livy's account of the expedition, until he brings the Car- 
 thaginians to the district called the Island. The river 
 designated by Livy the Arar, Bisarar, or Isara, is no doubt 
 the modern Isere. Hannibal, having crossed the Ehone 
 near Orange, would find himself, after his passage, in the 
 country occupied in the Roman times by the Cavares, and 
 a few miles from the southern frontier of the Tricastini. 
 Marching subsequently up the Rhone through the country 
 of the Tricastini and Sego-vellauni or Segalauni, he would 
 arrive at the Isere, the boundary of the district called the 
 Island. The Segalauni are not mentioned by Livy : they 
 are perhaps included by him among the Tricastini. Both 
 these tribes are included by Strabo among the Cavares » : it 
 is by Ptolemy that they are mentioned as distinct tribes 2. 
 These three tribes occupied the level country along the 
 Rhone from the Durance to the Isere. On their east? and 
 extending in like manner from the Durance to the Isere, 
 the Vocontii lived among the Alps bordering on the plain. 
 
 ^ Lib. IV. 
 
 2 Lib. n. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 131 
 
 Vaison^ and Die 2 were two of their towns ; and their terri- 
 tory along the Durance reached, according to Strabo 3, as 
 far as Embrun, where the Cottian land began. Among 'the 
 Vocontii must therefore be comprised a tribe called the 
 Caturiges, one of whose towns was Caturiges, or Caturigse, 
 now Chorges. It is necessary to distinguish this tribe 
 carefully from the Cottian Caturiges, who bordered on the 
 Salassi. Some other small tribes may also have been in- 
 cluded among the Vocontii. The eastern, or north-eastern, 
 boundary of the Vocontii seems to have been the river 
 Drac. Their frontier could not well have been far from 
 that river, on whichever bank it may have lain. 
 
 Before proceeding further with the consideration of 
 Livy's narrative, it will also be advisable to define, in this 
 place, the position of two other tribes, the Medulli and 
 Tricorii. The Medulli are mentioned in the list, given by 
 Pliny, of the Alpine tribes subdued by Augustus (Hist. 
 Nat. III. 20.)* Their name occurs there between the Acita- 
 vones and the Uceni. JSTow the Acitavones seem clearly to 
 be the same tribe as the Centrones. For the name of the 
 Centrones does not appear in the list, and the name of the 
 Acitavones is found where that of the Centrones, who 
 dwelt in the Tarentaise, should occur; (i.e.) after the 
 Nantuates, Seduni, Veragri, and Salassi. The Uceni in- 
 habited the Pays d'Oysans, the valley of the Romanche. 
 The Medulli would thus naturally be sought between the 
 Tarentaise and the Pays d^Oysans, (i.e.) in the Maurienne. 
 According to Ptolemy^, they lived above the AUobroges, 
 a statement which would also fix them in the Maurienne, 
 or, at any rate, the Lower Maurienne, as being the only 
 district bordering on the territory of the AUobroges not 
 
 1 Ptol. Lib. IL 2 Dea Vocontiorum. 3 Lib. IV. 
 
 4 Gentes Alpinae deylctae: Triumpilini, Camuni, Vennonetes, Isarci, 
 Breuni, Naunes, Focunates; Vindelicorum gentes quatuor, Consuanetes, 
 Viruclnates, Licates, Catenates, Abisontes, Rugusci, Suanetes, Calucones, 
 Brixentos, Lepontii, Viberi, Nantuates, Seduni, Veragri, Salasi, Acita- 
 vones, Medulli, Uceni, Caturiges, Brigiani, Sogiontii, Ebroduntii, Nema- 
 loni, Edenates, Esubiani, Veamini, Gallitse, Triulatti, Ectini, Vergunni, 
 Eguituri, Nementuri, Oratelli, Nerusi, Velauni, Suetri. 
 
 * 5 Lib. IL 
 
 9 — 2 
 
132 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 133 
 
 occupied by some other tribe. We should be led also, 
 from Strabo's account ^ to place the Medulli in the Mauri- 
 enne. * Above the Cavares,' Strabo says, * are the Vocon- 
 tii, the Tricorii, the Iconii, the Medulli.' He states again 
 further on ; ' Beyond the Vocontii are the Siconii, the Tri- 
 corii, and beyond these the Medulli, who inhabit the most 
 lofty mountains.' This description again tends to fix the 
 Medulli in the Maurienne, the Iconii or Siconii being pro- 
 bably the Uceni. The position of the Medulli being thus 
 ascertained, that of the Tricorii is easily found. They 
 dwelt between the Vocontii and the Medulli, bordering 
 also, in all probability, upon the Uceni. Their territory 
 would thus lie to the east of the Drac, and would extend 
 towards the Lower Maurienne and the Pays d'Oysans. 
 How far they reached to the south is uncertain ; perhaps 
 about as far as the present boundary of the department of 
 the Isere. 
 
 The Medulli, then, were the inhabitants of the Lower 
 Maurienne 2. The Upper Maurienne, it will afterwards 
 appear, was occupied by another tribe. The Tricorii dwelt 
 to the south-west of the Medulli, and to the south of the 
 
 1 Lib. IV. 
 
 2 The mountains of the Maurienne are very high ; a century ago, they 
 were thought, erroneously, to bo the highest of the Alps. Bilsching says 
 that "in Maurienne, more than in any other part (i.e. of the Alps) are 
 found mountains of immense height." This agrees with what Strabo 
 says of the position of the Medulli. 
 
 Strabo speaks also in this place of a pass from the country of the 
 Medulli into that of the Taurini ; but his account is very deficient in 
 accuracy. The pass, however, which he speaks of, seems clearly to be 
 that of the Little Mont Cenis. The length of the ascent, he says, is 
 stated to be 100 stadia ; and the descent to the frontiers of Italy is said 
 to be of the same length. Now the valley of the Arc is left, and the 
 ascent to the Little Mont Cenis begins, at Bramans ; from which place 
 to Susa, the frontier town of Italy, (it is named as such in the Itine- 
 rary from Bourdeaux to Jerusalem) the distance is very nearly 200 
 stadia, the lake on the summit of the Mont Cenis being about half-way. 
 Indeed, as will afterwards be shewn, the distance from Bramans to Susa, 
 by the Little Mont Cenis, is given in the Peutingerian Table as 25 Roman 
 miles, or exactly 200 stadia. 
 
 The great lake which Strabo mentions must be that on the Mont 
 Cenis. The exaggeration, by which this small lake is spoken of as a 
 
 trJ, 
 
 country of Allevard, and extended probably to Grenoble 
 and the river Drac. 
 
 Having followed up the Ehone, and arrived on the 
 banks of the Isere, Hannibal reached the confines of the 
 district called the Island. Livy's account of the trans- 
 actions in this country rather differs from that of Polybius. 
 According to Livy, Hannibal was chosen as arbitrator be- 
 tween the two brothers who were contending for the 
 sovereignty; while Polybius states that the assistance of 
 the Carthaginians was requested by the elder brother alone, 
 and that the younger brother was expelled in consequence 
 of Hannibal's joining his forces with those of the elder. 
 The discrepancy, however, is of no importance with regard 
 to the present question. 
 
 Livy calls the inhabitants of the Island, AUobroges. 
 They were such in his time, but not in that of Hannibal, 
 as is perfectly clear from the narrative of Polybius. Livy 
 indeed says at first, with respect to the Island, that the 
 AUobroges live near it. He should have said that they 
 lived near it in the time of Hannibal. It was subsequent 
 to that period that the Island was comprised in the ter- 
 ritory of the Allobroges, their name being probably applied 
 in the course of time to all the tribes they absorbed in 
 
 succession 
 
 1 
 
 great one, may lead us to infer that Strabo was not personally acquainted 
 with this country. Gray, however, in his letters, speaks of the lake on the 
 Mont Cenis as a * great lake/ Considering its position, it perhaps is so. 
 It is also a great inaccuracy to speak of the sources of the Durance and tho 
 Dora Baltea, the Salassian Dora, as being near one another, and on the 
 same mountain as the lake. Strabo has probably confounded with one 
 another the Dora Susina and the Dora Baltea, and combined together, 
 as if the two were but one pass, the features of the Mont Cenis and 
 the Mont Genevre. As both these roads cross the Cottian Alps, unite at 
 Susa, and lead into the country of the Taurini, the mistake may have 
 occurred. A similar error, as will afterwards appear, is made in the 
 Peutingerian Table, the * Alpis Cottia' and the * Alpes Cottise' being there 
 confounded. It is perhaps from this lake, the ' Ceann-uisg,' or 'river- 
 head,' that the pass has taken the name of the Mont 'Cenis.' 
 
 1 Something analogous to the probable extension of the name of the 
 Allobroges may perhaps be found in the case of the modern Switzerland, 
 the district so called originally comprising scarcely more than one-half of 
 the present Canton of Schwyz, 
 
134 
 
 han:nibal's passage of the alps. 
 
 The account which Livy gives of the march of Han- 
 nibal from the Island to the Alps is attended with con- 
 siderable difficulty : indeed, as previously mentioned, it is 
 inexplicable without alteration. The passage runs thus: 
 • Sedatis certaminibus Allobrogum, cum jam Alpes peteret, 
 non recta regione iter instituit, sed ad Icevam in Tricastinos 
 flexit ; inde per extremam oram Vocontiorum agri tendit 
 in Tricorios, hand usquam impedita via priusquam ad 
 Druentiam flumen pervenit.' 
 
 When Hannibal had arranged the affairs of the Island, 
 (the inhabitants of which Livy here erroneously calls Allo- 
 broges,) he would resume his march to the Alps, as we 
 have previously concluded from Polybius, from the con- 
 fluence of the Rhone and the Isere. But if from this point 
 he turned to the left, he would be going away from, 
 instead of approaching, the country of the Tricastini! 
 Besides, if Hannibal struck into the territory of the Tri- 
 castini, and then proceeded along the Vocontian frontier 
 till he reached the Tricorii, his movements would be quite 
 inexplicable. He would, on this supposition, have retraced 
 his steps down the Rhone, or nearly so, and then have 
 gained the country of the Tricorii by the very circuitous 
 route of Vaison, the Durance, and the valley of the Drac : 
 for it was along the Vocontian frontier, and not through 
 the Vocontian territory, that Hannibal directed his march 
 towards the Tricorii. It is quite clear that this route is 
 inadmissible, and that Livy's account is erroneous. 
 
 A single correction, however, grounded on the con- 
 clusions previously drawn from Livy's character, seems to 
 remove the difficulty. All the facts in the passage cited 
 are true, but one of them is misplaced. The transactions 
 at the Island occurred, not before, but after, the passage 
 through the country of the Tricastini. When Hannibal 
 had crossed the Rhone, he had the Alps immediately before 
 him : their passage was the next act of the expedition : 
 then it was ' cum jam Alpes peteret.' Now had he made 
 directly for the mountains, he would have gone to the east. 
 This, however, he did not do: *non recta regione iter 
 instituit.' Instead of proceeding in this direction, Han- 
 nibal, being then near Orange, almost on the northern 
 frontier of the Cavares, ' ad laevam in Tricastinos flexit,' 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 135 
 
 f 
 
 turned to the left (the north) into the country of the Tri- 
 castini, and thus arrived, marching along the side of the 
 Rhone, at the point where the Isere joins it. The disputes 
 of the men of the Island were then settled, and the march 
 resumed from the confluence of the two rivers. * Inde per 
 extremam oram Vocontiorum agri tendit in Tricorios,' he 
 directed his march along the left bank of the Isere, the 
 northern boundary of the Vocontian territory, and thus 
 reached the country of the Tricorii, who dwelt beyond the 
 Drac. This seems to be the course indicated by Livy as 
 that which Hannibal followed. It is the same as that which 
 has been deduced from Polybius, and is clearly the only 
 route by which' the statements of both historians can be 
 reconciled ^ 
 
 Another difficulty now results from Livy's narrative. 
 He speaks of the Durance (Druentia) as lying on Hanni- 
 baPs way to the Alps. This, however, is not possible. In 
 order to reach the Durance from the neighbourhood of 
 Grenoble, Hannibal must have followed up the valley of 
 the Drac, crossed the mountains to Gap, and thus arrived 
 at the Durance. But, after crossing the Durance, Livy 
 says that Hannibal's route lay through a country of a level 
 character, which certainly would not be found beyond the 
 Durance, or in the valley of that river. Neither could 
 Hannibal arrive by this route in the country of the Allo- 
 broges, the first tribe, according to Polybius, by whom 
 Hannibal was attacked. Livy does not, it is true, call this 
 tribe Allobroges, but Polybius's authority is conclusive upon 
 the subject, and Livy speaks of the attack as being subse- 
 quent to the passage of the Druentia. 
 
 There seems to be little doubt that Livy has made 
 another error here, and has mistaken some river of similar 
 name for the Durance; and, under the influence of this 
 deception, has introduced the description he gives of the 
 Durance, of which river it is a sufficiently correct account. 
 
 1 It is conjectured by the author of A Dissertation on the passage of 
 Hannibal over the AlpSf by a member of the University of Occfordi (p. 95), 
 that the text of Livy is corrupt in this place. There seems however no 
 BuflScient reason for adopting this supposition. The fault is Livy's, 
 rather than his transcriber's. The text is probably correct, but the his- 
 torian has fallen into an error. 
 
136 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS 
 
 Now the name Durance, a term of Celtic origin, is applied, 
 under different forms, to more than one Alpine river ^ In 
 the Chablais and the Vallais, it occurs under the con- 
 tracted form Dranse, a name so nearly resembling Drac, 
 as to render it not improbable, that it was the Drac which 
 was mistaken by Livy for the Durance 2. The Eoman name 
 of the Drac is not known: but, as the word Drac, or 
 Dur-ac, is unquestionably Celtic, the river must have borne 
 among its inhabitants, in the Iloman times, a name not 
 greatly different from that which it bears at present, and 
 therefore one having a considerable resemblance to Dranse 
 and Durance. It is thus not unreasonable to suppose, 
 that, misled by the similarity of names, Livy mistook, in 
 the accounts given him, the Drac for the Durance, called 
 in consequence the Drac, Druentia, and applied to it the 
 description he gives, which he no doubt intended for the 
 Durance. It will be seen at once, on reference to the 
 map, that an army, marching up the left bank of the Isere 
 into the country of the Tricorii, would find the Drac 
 directly in their way. This river is at all times a con- 
 siderable mountain-stream: when swollen with rain, as 
 Livy states it to have been when Hannibal crossed it, 
 the passage would have presented sufficient difficulty to be 
 of the importance he assigns to it^. 
 
 Having crossed the Drac, Hannibal's future march 
 must have lain up the Graisivaudan ; for, according to 
 Livy, the country through which he passed was of a level 
 character. * Hannibal ab Druentia campestri maxime inti- 
 nere ad Alpes cum bona pace incolentium ea loca Gallo- 
 rum pervenit.' This cannot be said of the valley of the 
 Drac or of the valley of the Romanche, through which lie 
 the two routes to the Mont Genevre. It was not then the 
 Mont Genevre which Hannibal crossed; but, as far as 
 appears from this portion of the narrative, the Mont Cenis, 
 or else the Great or Little St Bernard, all which passes 
 
 1 The first syllable is the word « dur' or * dwr,' ' water/ The second 
 element is found in the Anza, the name of the river of the Val Anzasca. 
 
 2 Strabo calls the Dora Baltea, * Apovemia,' Lib. V. 
 
 3 The Drac, and the Druentia of Livy, have been previously concluded 
 to be identical by M. Larauza. See Ukert's Geographic, Appendix on 
 HannibaVs Passage of the Alps, 
 
 137 
 
 
 might be reached by the way of the Graisivaudan. But 
 Hannibal eventually arrived, after crossing the Alps, ac- 
 cording to Livy, as well as Polybius and Strabo, in the 
 country of the Taurini ; so that he must have passed the 
 Mont Cenis, and not either of the St Bernards, both of 
 the latter roads leading through the country of the Salassi 
 into that of the Libui or Libicii. 
 
 After noticing the Tricorii, no proper names occur 
 in Livy till the Taurini are mentioned. The historian loses 
 himself among the Alps, a region of which he appears not 
 only personally ignorant, but also to have failed in forming 
 any tolerably accurate conception. Having brought the 
 Carthaginians to the foot of the Alps, he speaks as if the 
 view of the mountains had then for the first time burst 
 upon them, and transplants the features of the high Alps 
 to the country at the commencement of the mountains, 
 which would have been of an entirely different character. 
 *Tum, quanquam fama prius, qua incerta in majus vero 
 ferri solent, praecepta res erat, tamen ex propinquo visa 
 montium altitudo nivesque coelo prope immixtse, tecta 
 informia imposita rupibus, pecora jumentaque torrida 
 frigore, homines intonsi et inculti, animalia inanimataque 
 omnia rigentia gelu, caetera visu quam dictu foediora, ter- 
 rorem renovarunt.' This description, eloquent and pic- 
 turesque as it is, and however graphic it may still be^ 
 when applied to the high valleys in the Alps, is quite out 
 of place where it stands. 
 
 Livy's account of the contest with the AUobroges, and 
 the attack made upon Hannibal at the * strong white rock ' 
 of Polybius, may be passed over. Polybius has clearly 
 furnished here the main groundwork of the Latin narrative : 
 and it would have been well if Livy had adhered more 
 closely than he has done to his principal authority. For his 
 accounts of both battles are very confused, and cannot be 
 correct. His explanation of the separation of the Cartha- 
 ginian army into two parts near the Rock is exceedingly 
 faulty, and quite at variance with the narrative of Poly- 
 bius, though clearly drawn from it. Livy does not seem to 
 
 ^ See Gray's Letters, 
 
138 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 have understood how the separation of the army into two 
 bodies could have ensured its safety. He therefore aban- 
 dons his authority, and makes the Carthaginian line of 
 march cut in two by the Gauls, who descend from the 
 mountains into the valley, interpose themselves between 
 the two divisions of the army which they have thus 
 broken through, and remain in this position all night. Yet 
 from this important stroke, which must inevitably have 
 brought on a severe contest on the following day, nothing 
 results ; and the two divisions of the Carthaginian army 
 are represented as being reunited on that day with little 
 or no difficulty, the opposition of the Gauls being much 
 relaxed. The heat of the conflict thus subsides, according 
 to Livy, exactly at the time when it ought, as far as can be 
 judged, to have risen to the highest pitch \ 
 
 There must be considerable exaggeration in what Livy 
 relates, of the Carthaginians having been led astray by 
 their guides, and of their endeavouring to find a way for 
 themselves, by following up certain valleys at random. The 
 Cisalpine Gauls, who were with Hannibal, would have been 
 able to act as guides to the Carthaginians sufficiently well 
 to prevent them making any important deviations from 
 the right course. In certain local difficulties they may, 
 however, have had recourse to the inhabitants of the 
 country, and have pressed some of them into their service 
 as guides. By such guides, if they employed them, they 
 might have been led a little out of their way in places ; 
 but that they should have gone astray to any important 
 extent is not likely. As to their seeking a way for them- 
 selves by entering into different valleys on conjecture, the 
 supposition is very improbable, and can hardly be enter- 
 tained for a moment. 
 
 Livy's account of HannibaPs exhibiting to his soldiers 
 from the summit of the pass, the prospect of Italy, seems 
 partially drawn from some other source than Polybius 
 
 1 Livy speaks of the first meeting with these Gauls, and their subse- 
 quent attack upon Hannibal, as if both events occurred on the same day 
 (i.e.) the fourth day after leaving the * town/ Ho omits the two days of 
 guidance mentioned by Polybius. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 139 
 
 and presents us with some additional circumstances. Two 
 days, Livy says, the Carthaginians remained encamped 
 at the summit of the pass. With this circumstance Poly- 
 bius supplied him: what next follows appears to have 
 been derived from another quarter. At the first dawn of 
 day (the narrative proceeds) the standards were advanced, 
 (apparently leading the way down into Italy). The marching 
 columns followed slowly and without spirit through the 
 snow, while an expression of sluggish and hopeless de- 
 jection was stamped on the faces of all the men. Han- 
 nibal, seeing this, passed on before the standards, and 
 ordered a halt to be made at a certain promontory, which 
 commanded a very extensive view, and from whence he 
 pointed out Italy to his soldiers ^ * Per omnia nive oppleta 
 cum signis prima luce motis segniter agmen incederet, 
 pigritiaque et desperatio in omnium vultu emineret, prse- 
 gressus signa Hannibal in promontorio quodam, unde longe 
 ac late prospectus erat, consistere jussis militibus Italiani 
 ostentat subjectosque Alpinis montibus Circumpadanos 
 campos, moeniaque eos tum transcendere non Italise modo 
 sed etiam urbis Romanae.' 
 
 The place from which it has been concluded that Han- 
 nibal pointed out Italy to his soldiers, has been already 
 mentioned^ It is a narrow ridge about half a mile long, 
 nearly level at the summit, and separated on the north 
 from the rest of the hilly range to which it belongs by the 
 hollow through which runs the road from the Little Mont 
 Cenis. It presents a very steep slope towards the plain 
 of La Grande Croix on the east, and terminates on the 
 south, above the plain of St Nicholas, in a sheer preci- 
 pice many hundreds of feet in height. An eminence of this 
 kind is accurately defined as a promontory : and it is very 
 remarkable, that the consideration of Polybius' narrative 
 alone, who is quite silent as to the nature of the point 
 from which Italy was seen, should have led to the con- 
 clusion, independently of Livy, that it was from this pro- 
 
 1 This circumstance renders the adoption of any pass but the Mont 
 Cenis irreconcileable with Livy's narrative. 
 
 2 P. 115. 
 
140 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 montory that Hannibal pointed out to his soldiers the 
 plains of the Po. 
 
 According to Polybius, Italy was seen while the Car- 
 thaginians remained on the summit of the pass. According 
 to Livy, it would appear that it was seen at the com- 
 mencement of the descent. The conclusion, which has 
 been drawn, that it was seen from the road, or close to 
 the road, between the plateau of the Little Mont Cenis 
 and La Grande Croix, reconciles this apparent disagree- 
 ment. For the descent from this pass into Italy is con- 
 sidered to begin from La Grande Croix, all the ground 
 above, though varying considerably in level, being included 
 m the plateaux which form the summit of the Mont Cenis. 
 Hannibal, therefore, when on the promontory, or at La 
 Grande Croix, would still be correctly spoken of by Poly 
 bius as being on the summit of the pass. Yet he mio-ht 
 naturally be mentioned by another author as having begun 
 his descent, when on his way from the plateau of the Little 
 Mont Cenis to La Grande Croix. For La Grande Croix is 
 600 feet below the Col of the Little Mont Cenis, and not 
 less than 400 below the general level of that plateau ; so 
 that the route from the plateau of the Little Mont Cenis 
 to La Grande Croix is, strictly speaking, a descent. It 
 would be very singular, if the Mont Cenis were not the 
 pass crossed by Hannibal, that it should afford so obvious 
 and simple an explanation of what at first sight appears 
 a direct contradiction between the two authors. 
 
 Livy has not observed, that, if the day on which Italy 
 was seen be included among the days of descent, two days 
 cannot be allowed for the encampment at the summit of 
 the pass. He should therefore have diminished to one the 
 two days he mentions, and which he found given in 
 Polybius By these means he would have made his narra- 
 tive of the events at the summit of the pass quite con- 
 sistent m Itself, and completely in accordance, though 
 at first sight apparently at variance, with the account of 
 the Greek historian. 
 
 Livy's statement, that the descent from the Alps on 
 the side of Italy, is shorter and more rapid than it is on 
 the other side, is quite correct. On the pass of the Mont 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 141 
 
 \ 
 
 Cenis this is particularly remarkable; Susa being more 
 than 5000 feet below either of the Cols of the Great or 
 Little Mont Cenis. 
 
 In relating the events which occurred at the place 
 where the path was destroyed, Livy has introduced some 
 additional circumstances. It was by the falling away of 
 the earth, Livy says, that the path was broken up. This 
 is no doubt correct : but when he mentions that the effect 
 of this landslip was to produce a precipice a thousand feet 
 high, he has plainly fallen into an error by misinterpreting 
 Polybius. The broken ground, along which the path ran, 
 was a thousand feet across : the measurement has nothing 
 to do with the height of the precipice, which the path 
 skirted. 
 
 This precipice, along which the path lay, was, accord- 
 ing to Livy, so steep, that a light-armed soldier could only 
 manage with difficulty to accomplish the descent, by avail- 
 ing himself of the bushes and stumps which presented 
 themselves here and there upon the declivity. The moun- 
 tain-side between La Ferriere and La Novalese, which has 
 been previously mentioned as the place where Hannibal 
 was most probably stopped, is a precipice of exactly this 
 kind. There are several bushes scattered about its sur- 
 face, by which an unencumbered man would be greatly 
 assisted in an attempt to get down, and by the aid of 
 which he might succeed, without much difficulty, in effect- 
 ing a descent. 
 
 The ' glacies ' of Livy is identical with the ' vetus nix ' 
 he mentions just before. It was old snow, which had be- 
 come hard and solid, and not real ice. It might be merely 
 the want of a word to express this old snow, which induced 
 Livy to call it ' glacies ' or * ice^* 
 
 1 In Smollet*s Travels through France and Itali/, lietter xxxYU.f he 
 speaks of the hard snow covering in March the Col de Tende as * ice.' 
 Latrobe, also, in his Alpenstock, speaks of * ice' in summer on the Great 
 St Bernard. Ice is likewise spoken of on the passes of the Spliigen, 
 and of the Monte Tonale, in the account of Macdonald's expedition in 
 the year 1800. (Alison's Europe^ Chap, xxxii.) Indeed, in this latter 
 work, even glaciers are mentioned as existing on passes perfectly free 
 from them, as on the Susten, the Niifanen, and the Wormser Joch. 
 
142 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 Hannibal, having been defeated in his attempt to effect 
 a circuitous route, by the difficulties which the old snow 
 presented, was at length obliged to encamp above the 
 broken path. 
 
 'Tandem nequicquam jumentis atque hominibus fati- 
 gatis, castra in jugo posita, aegerrime ad id ipsum loco 
 purgato: tantum nivis fodiendum atque egerendum fuit 
 Inde ad rupem muniendam*, per quam unam via esse po- 
 terat, milites ducti, cum caedendum esset saxum, arboribus 
 circa immanibus dejectis detruncatisque struem ingentem 
 lignorum faciunt, eamque, cum et vis venti apta faciendo 
 igni coorta esset, succendunt, ardentiaque saxa infusa aceto 
 putrefaciunt. Ita torridam incendio rupem ferro pandunt, 
 moUiuntque amfractibus modicis clivos, ut non jumenta 
 solum sed elephanti etiam deduci possent ' 
 
 This story seems very extravagant. Polybius makes 
 no mention of this method of breaking up the rock, nor 
 even of cutting a way through the rock at all. Besides, it 
 is certain, both from Polybius and even from Livy himself, 
 that there were no great trees, nor even any trees whatever^ 
 in this neighbourhood. Nothing but some bushes would 
 have been found to make a fire with : the vast pile of trees 
 mentioned by Livy could never have been constructed. 
 And even if the materials could have been found for it, it is 
 difficult to say where it could have been placed on the' face 
 of a precipice, which seems to be the position Livy assigns 
 to it. The liquid mentioned as *acetum' or 'vinegar' 
 would probably be the thin acid wine, which the soldiers 
 may have had with them, and which might have been used 
 because there was little or no water to be found in that snow- 
 covered neighbourhood. They would hardly, however, have 
 had a sufficient quantity of this vinegar to have been em- 
 ployed in making a long zigzag route down a precipice, or 
 even perhaps to have broken up the rock to any extent. 
 
 Yet it is only snow which could have been found on any of these passes 
 On the Col de Tende and the Monte Tonale the snow is not even per- 
 manent through the summer. A word is in fact wanting for this old 
 and hardened snow : the use of the word *neve' is confined to the upper 
 and snowy parts of glaciers, and is not applied to hard snow generally. 
 
 ^ Tov Kprjfxvdp <fa)<oSo>«, Pol}b. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 143 
 
 Yet, although this story is clearly extravagant, it is not 
 satisfactory to reject it as a pure invention. Such tales 
 are generally exaggerations or perversions of something 
 true, and are seldom quite groundless. The foundation 
 of this story is, however, difficult to be discovered, or 
 its omission by Polybius to be accounted for. This omis- 
 sion, indeed, seems of itself to prove that it was not a 
 very important or singular event. The true explanation 
 of the origin of the story must always remain very doubt- 
 ful, and it seems impossible to offer a conjecture on the 
 subject with any confidence^ 
 
 1 There is perhaps some room for a suspicion, that it was not the rock, 
 but the old snow hardened almost to the consistency of rock, which was 
 cut through by Hannibal. The want of a proper word for old snow, by 
 which it might have been distinguished from fresh snow, may partially 
 have conduced to such an eiTor, by yitiating the story as it came down 
 from hand to hand till it reached Livy. The manner in which the event 
 is related by Appian gives some colour to this surmise. He says : * As 
 there was a great quantity of snow, and it had been rendered hard by 
 frost, (this seems the force of Kpvovs here) Hannibal cut some wood near 
 and burnt it, and then poured water and vinegar on the ashes : and the 
 rock being by these means decomposed, he broke it up with iron ham- 
 mers, and made a way.' x'^^^^^ ''^ ttoWtjs ovarjs koi Kpvovsy rrjv fiev vXrjp 
 rtfJLVwp T€ Koi KaraKaiojVf ttjv be T€(j)pav a^evj/vs vdoTi koi of^t, koi rrjv TTfTpav 
 fK TOvBe yjracfiapav yiyvofxepiju, a-(j)vpais (TiBTjpdis 6pav<ov Ka\ oBoTTOiav. — Appian, 
 Lib. VII. De Bello Annibalico. 
 
 There is of couree no doubt about Appian*s meaning here, the break- 
 ing up of the rock being plainly stated : indeed he immediately after- 
 wards expressly mentions a particular cutting in the rock as being known 
 by the name of the 8io8os *Apvi^ov ; a groundless tradition probably, but 
 alluded to in such a manner, as to make Appian's own opinion on the 
 subject perfectly clear. Still, though there can be no question about Ap- 
 pian's meaning in the passage quoted, its strikingly inconsequent character 
 tends to raise a doubt as to his right apprehension of the accounts from 
 which he composed his narrative. For there could be no reason why 
 Hannibal should have broken up the rock, on account of the presence 
 of snow solidified by frost : and yet we can hardly help inferring from 
 Appian's words, that it was this snow which was the cause of the cutting, 
 in whatever substance that cutting may have been made. If, however, 
 the words ttjp nerpap could be referred to the solid snow, all would be 
 easy and consistent. Hannibal found his way obstructed by it in some 
 place, and was obliged to cut through the hardened mass, decomposing it 
 
141 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 According to Livy, four days were spent in making the 
 new road. This error arises from his misinterpreting 
 Polybius, as will easily be seen on reference to that histo- 
 
 first by the aid of fire, and the percolation of some warm liquid, and thus 
 enabling himself to break it up with greater facility. 
 
 Instances are found in modern Alpine campaigns of roads being cut 
 through hard snow. In the winter campaign of Macdonald in the 
 Bhsetian Alps, the best illustrative parallel to Hannibal's expedition, two 
 cases are mentioned. The first relates to the passage of the Splugen. 
 •In the narrow plain between the glaciers, (i. e. on the summit of the 
 pass, beneath the higher glacier-bearing peaks of the ridge crossed) 
 however, they found the road blocked up by an immense mass of snow, 
 formed by an avalanche newly fallen ; upon which the guides refused to 
 enter, and in consequence the soldiers returned, unanimously exclaiming 
 
 that the passage was closed Put to shame by such an example, the 
 
 troops and peasants redoubled their efforts; the vast walls of ice and snow 
 were cut through, and although the hurricane increased with fearful rapid- 
 ity, and repeatedly filled up their excavations, they at length succeeded 
 in rendering the passage practicable.'— Alison's Hist, of Europe, iv. 431, 
 432, 5th Edit. Jomini, however, says nothing of ice or avalanches in this' 
 place, but only speaks of the great depth of the snow on the plateau at 
 the summit of the pass, and of the fresh snow continually falling, (i. e. 
 from the sky). In the account of the passage of the ridge of San Zeno,' 
 a pass among the low mountains between the lakes of Iseo and Idro,' 
 bordering on the Lombard plain, another instance occurs. * The diffi! 
 culties of the ridge of San Zeno, however, had almost arrested the 
 soldiers whom the snows of the Splugen had been unable to overcome • 
 a few horses only could be got over by cutting through blocks of id 
 (compact snow) as hard as rock at the summit, (apres avoir taille un 
 chemm dans la glace vive, Jomini) and the greater part of the cavalry 
 and artillery required to descend by the smiling shores of the Lago Isea 
 (Lago dlseo) to Brescia, and ascend again the vine-clad banks of the 
 Chiesa. — Alison's History of Europe, iv. 446. 
 
 On the declivity between La Ferriere and La Novalese, where Hanni- 
 bal 8 path was broken up, he may possibly have been obliged to cut 
 through the mass of solid snow, which, it has been conjectured, filled at 
 that time the gully which penetrates into the mountain-side. For, after 
 the path had been reconstructed along the precipice above (i. e. to the 
 west of) the gully, it would then have been necessary to cross the gully 
 Itself. But the snow, however solid, which filled the gully, would pro- 
 ^bably not have been capable of supporting the elephants; for even the 
 beasts of burden had become fixed in it, when the circuitous route was 
 attempted It might consequently have been necessary to make a lane 
 through the snow, and perhaps to penetrate to the ground beneath. Yet 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 145 
 
 rian. Had Livy attended to the words iv ^/uepaig rpial 
 KaKowaOi^aa?, he would not have fallen into this mistake, 
 which would lead us to imagine that the first day spent in 
 the midst of the snow was one of comfort, and that the 
 hardships the army suffered only began on the second 
 day, since there were but three days of hardship in all. 
 
 The three days (which must be reduced to two) during 
 which the men and baggage-animals, (the cavalry horses 
 should be included) enjoyed an interval of rest, and were 
 enabled to recruit their strength a little, were not, as 
 
 this lane would not have been long, as the gully is not wide, and as it 
 was probably only in the gully that the old snow lay. 
 
 An operation of this nature would not have been of any great 
 importance, and might have been passed over by Polybius, or not have 
 come to his knowledge. That he had received no such intimation is 
 perhaps most likely. He does, however, when he first speaks of the 
 paxis, (a word he introduces as if he had mentioned an equivalent for it 
 before) employ the expression diafirja-afievos ttjp cV avrfj ^toVa. Now, if 
 paxis could mean the side of a ridge or mountain, the words diafirja-a^ 
 fxevos, K. T. X. might signify * clearing a way through the snow upon it.' 
 The proper meaning of pa^t? is, however, the top, and not the side of a 
 ridge ; and the removal of the snow is spoken of in connexion with the 
 encampment irepl ttjv paxt-v, and cannot well bo supposed to refer to 
 the construction of the path. It is true, indeed, that Polybius' informa- 
 tion may have been somewhat defective on this point, or his compre- 
 hension of what was related to him not quite perfect. In his account of 
 the encampment at the summit of the pass, where there must have been 
 snow, nothing is said of its removal. 
 
 If Hannibal really cut a path through the solid snow, it may possibly 
 have been efiected in the following manner. 
 
 The workmen, perhaps, first levelled the surface of the snow a little, 
 on the line of the intended cutting; and then, collecting the bushes 
 near, arranged them along it, and set them on fire. The surface of the 
 snow would thus be partially melted, and the water produced would 
 sink down through the snow beneath. A quantity of liquid thrown on 
 the hot ashes would also have helped to dissolve this snow to some 
 extent. By these means, the mass, originally hard and firm, would be 
 decomposed and softened on the line of the projected path, and might 
 have been broken up with pickaxes or similar instruments without great 
 difficulty. To cut a way through old and solid snow, without taking any^ 
 measures to soften it, would be a tedious and laborious task. 
 
 This supposition may possibly explain Livy's story. It must, however, 
 be looked upon as nothing more than a mere conjecture. 
 
 10 
 
146 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 might be at first supposed from Livy's account, days 
 subsequent to the passage of the elephants, but the two 
 days which succeeded the passage of the horses and bag- 
 gage-animals, and during which the elephants were de- 
 tained in the higher encampment. 
 
 Nothing is related by Livy as having occurred, after 
 the new road was completed and passed, during the rest of 
 the descent to the plains of Italy. Hannibal there entered 
 into the country of the Taurini, a fact which Livy says was 
 universally agreed upon. He then proceeds to mention 
 two passes by which it had been erroneously supposed that 
 Hannibal crossed the Alps : the Pennine Alp or Great St 
 Bernard, and the ' Cremonis jugum,' probably the Little 
 St Bernard, the ancient name being apparently preserved 
 in the neighbouring peak of the Crammont. Neither of 
 these, as Livy observes, could have been crossed by Han- 
 nibal, as they both lead, not into the country of the 
 Taurini, but through that of the Salassi into the territory 
 of the Libui. The name of the Salassi, indeed, although a 
 large tribe, is never found at all in Polybius; and'^the 
 name of the Libui, although mentioned by him elsewhere 
 does not occur in his account of HannibaPs passage of 
 the Alps. 
 
 V 
 
 CHAPTER X. 
 
 Examination of objections brought against the Mont Cenis. 
 
 WE have now examined the narratives of Polybius and 
 Livy, the two chief historians of Hannibal's expedi- 
 tion. Both seem to point to the Mont Cenis, as the pass 
 by which the Carthaginian general effected his descent into 
 Italy. To complete the evidence in favour of this pass, 
 it will now be requisite to take some notice of the objec- 
 tions which have been brought against it. 
 
 Nearly all the objections which can be adduced, are 
 collected together, in his Histoire du passage des Alpes par 
 Annihal, by M. Deluc, one of the advocates for the pass of 
 the Little St Bernard. The sixth chapter of the second 
 book of his work is entitled a *Eefutation des auteurs 
 qui conduisent Annibal par le Mont Cenis,' and in it we 
 find the following summary of the objections against this 
 pass. 
 
 * Ce qu'il y a de tres-remarquable dans la premiere 
 route (the route of the Little St Bernard) c'est qu'elle ne 
 traverse pas une seule fois ITsere, la vallee n'etant nuUe 
 part aussi resserree pour forcer a etablir des ponts sur 
 cette riviere ; dans la vallee de TArc, au contraire, il y a 
 un si grand nombre d'obstacles, que Ton est force de 
 passer dix fois d'une rive a Tautre pour les eviter. " II 
 serait trop long de detailler," dit M. de Saussure, "les 
 nombreux defiles que Ton passe dans cette route, et de 
 noter combien de fois les etranglemens de la vallee et les 
 sinuosites de TArc forcent k passer d'une rive a Tautre." 
 
 Cette vallee offrait done de trop grandes difficultes 
 pour que, dans les temps recules, on y etlt fait passer 
 une route pour traverser les Alpes. La descente du Mont 
 Cenis, du cote de ITtalie, etait aussi un trop grand obstacle, 
 car les rochers y sont presque a pic, et ce n'est qu'en 
 taillant le chemin dans le roc avec un grand nombre de 
 zigzags, qu'on a pu rendre cette descente praticable. 
 
 10—2 
 
]48 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 C'est sans cloute a cause de ces difficultes naturelles 
 que la route du Mont Cenis n'a ete ouverte que dans les 
 temps modernes, compares a Tanciennete de la route du 
 Petit Saint-Bernard ; aussi la premiere ne se trouve point 
 dans les itineraires romains, qui, cependant, ont ete faits 
 dans le 4^ et 5« siecles de notre ere, ou six a sept siecles 
 apres Texpedition d'Annibal. 
 
 La route du Mont Cenis n'etait done pas celle que les 
 Gaulois suivaient pour descendre en Italic, ni celle qu'An- 
 nibal, en marchant sur leurs traces, prit pour entrer dans 
 le meme pays ; nous ferons aussi observer qu'elle n'etait 
 pas une des quatres routes qui seules etaient connues du 
 temps de Poljbe. Cet auteur, en decrivant la route d'An- 
 nibal, la meme qu'il parcourut cinquante ou soixante 
 annees apres, ne pent decrire qu'une route qui etait connue 
 de son temps. 
 
 Nous observerons encore qu'a la descente du Mont 
 Cenis, il est impossible qu*on pAt rencontrer, a la fin d'octo- 
 bre, de la vieille neige conservee depuis I'hiver precedent ; 
 car, outre que ce passage est plus abaisse d'au moins 100 
 toises que celui du Petit Saint-Bernard, sa descente est 
 tournee vers le sud-est, exposition ou la neige fond plus 
 vite que dans celle du Petit Saint-Bernard, qui est tournee 
 vers le nord est.' 
 
 The objections embodied above amount to the fol- 
 lowing : 
 
 (1) The passage of the Mont Cenis must have been 
 very difficult in the time of Hannibal. 
 
 (2) The road to the Mont Cenis crosses the Arc a 
 great many times. 
 
 (3) The descent into Italy is very precipitous and 
 difficult. 
 
 (4) The route of the Mont Cenis is not found in the 
 Roman Itineraries. 
 
 (5) It was not one of the four routes known to Po- 
 lybius. 
 
 (6) It would have been impossible for Hannibal to 
 have found old snow on the descent from the Mont Cenis. 
 
 (1) The fact of the passage of the Mont Cenis having 
 been very difficult in the time of Hannibal may be readily 
 
 7 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 149 
 
 granted. Such must inevitably have been the case with 
 every Alpine route in the Gallic period. Yet, although 
 the Mont Cenis must have been then a difficult pass, that 
 difficulty can scarcely be considered sufficient to disqualify 
 it for being received as the scene of the passage, de- 
 scribed as * toilsome, and excessively difficult, although not 
 impossible,' eiriTrovov julcv Kai cva")^€prj Xiav, ov fxijv a^vuaTou : 
 for such are the epithets which are applied in Polybius to 
 the pass which Hannibal crossed. Indeed, at the very 
 commencement of the Alpine route, we read of rd^ 
 Suaxt^p^oi^' Between the town of the AUobroges and the 
 ' strong white rock,' we are told again of rds e^rj^ Suax^pici^' 
 By the Rock itself, also, we read of a (pdpayyd Tiva Suafia- 
 Tov Kal KprjiJivciSrj, All these expressions certainly imply dif- 
 ficulty ; and, as to the descent into Italy, many thousands 
 of men were lost, viro twv tottwv kui t^£ yiovo^y in conse- 
 quence of the snow and the difficult or dangerous character 
 of the descent. 
 
 To shew, then, that the Mont Cenis was a difficult pass 
 in the time of Hannibal, proves nothing against it : for the 
 pass which Hannibal crossed was also difficult. In order 
 to exclude the Mont Cenis on this ground, it would be 
 necessary to bring against it an amount of difficulty insu- 
 perable in ancient times. But this certainly cannot be 
 concluded of a pass, which, however difficult, must rank, 
 relatively to Alpine passes generally, among the easiest 
 known ; a pass which may be shewn to have been always 
 a thoroughfare through the Alps, and which is now, what 
 it seems to have been for some centuries, the great high- 
 way from France into Italy. 
 
 (2) The road from the Valley of the Isere to the Mont 
 Cenis certainly crosses the Arc very often : the old road 
 to the little Mont Cenis would have passed the river six 
 times. Once also, on the descent into Italy, it would have 
 to cross the Dora Susina. But, although the modern 
 road from Chambery to the Little St Bernard never crosses 
 the Isere, yet the older road, the Roman way from Lemin- 
 cum, probably crossed that river four times, and the Are 
 once. For, of the Roman stations, Axima, Darantasia, 
 Oblimum, Mantala, and Lemincum, each lay, in all pro- 
 
150 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 bability, as will afterwards be shewn \ on a different 
 side of the Isere from the following station. Besides, on 
 the descent from the Little St Bernard, even the modern 
 road crosses the Dora Baltea three times : and it cannot 
 be conceived, why the Arc should have offered such a 
 great obstacle to HannibaPs progress, and the passage of 
 the Dora Baltea have presented little difficulty. 
 
 But, in point of fact, the passage of these mountain- 
 streams, at the end of October, would have been attended 
 with no great difficulty, unless, as in the case of the Drac, 
 they were swoln with rain. They would probably have 
 contributed but little to the losses, which Hannibal is said 
 by Polybius to have incurred in crossing rivers. For all 
 these torrents are fed by the melting of the snows and 
 glaciers, and present, during the long Alpine winter, no 
 resemblance to the full and turbid streams, which sweep 
 down the valleys in the summer months. At the end of 
 the month of September, the last summer month, (the 
 Alps, at least the high Alps, may be said to have no 
 autumn) the action of the sun on the snow and ice begins 
 rapidly to diminish in power, and the torrents to shrink in 
 volume. Throughout the winter, little or no water is sup- 
 plied by the glaciers to the mountain-rivers 2, which become 
 
 ^ See Note A. at the end of the book. 
 
 2 Si, comme je Tai montre par les tableaux du jaugeage de TAar, il 
 existe un rapport direct entre I'etat atmospherique et le debit des tor- 
 rents, il doit arriver un moment en automne, aH toutes les rivieres suhissent 
 urn baisse notable, h raison de la fonte moins considerable qui a lieu k 
 cette saison. Par la memo raison, elles doivent tarir completement en 
 hiver, du moment que toute fonte a cesse h la surface. Voulant m'en 
 assurer par I'observation directe je n'ai pas hesite h entreprendre dans ce 
 but un voyage hivernal dans les glaciers. Au commencement du mois 
 de mars 1841, par consequent h une epoque ou I'hiver regne encore 
 d*une maniere absolue dans les Hautes-Alpes, je me rendis avec M. Desor 
 au glacier de TAar que je remontai jusqu ^ THotel des Neuchatelois (le 
 12 Mars) [A great rock on the glacier of the Aar was called by this 
 name]. Apres avoir sejourne deux jours au Qrimsel, nous visitames 
 ensemble le glacier de Rosenlaui (le 14 Mars.) 
 
 A Meyringen deja je trouvai I'Aar beaucoup plus faible qu'en ete ; 
 h. mesure que je montai, je la vis diminuer toujours plus, si bien qu'^ 
 la Handeck clle etait reduito a un tres-petit filet d'eau, a peine com- 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 151 
 
 in general clear and shallow streams, with a comparatively 
 gentle current. At the end of the month of October, two 
 thousand years ago, Hannibal would have found no great 
 difficulty in fording the Arc, or any other of the Alpine 
 rivers. 
 
 (3) The precipitous character of the descent from the 
 Mont Cenis, brought by M. Deluc as an argument against 
 that pass, is in strict accordance with what is related by 
 Polybius with respect to Hannibal's descent from the Alps, 
 and is one of the distinctive features by which the Mont 
 Cenis is identified with the pass which Hannibal crossed. 
 Certainly, not only is the descent of the Fourmiguier pre- 
 cipitous, but, in order to make a path of commodious 
 width down it, it would most probably have been necessary 
 to cut the rock away in places. Nothing but a rugged 
 and narrow path could have existed, in barbarous times, in 
 such a situation. Yet this precipitous and narrow way is 
 precisely what we find mentioned by Polybius. Ouatjs ydp 
 (TTeviJ9 ffaJ KaT(i)(f)€pov9 rrj^ KaTa(ia(Tew^, t^s 5e ')^i6vo^ aStjXov 
 TTOiovo-rji €Kd(jToi9 Tfjv CTrif^aGLV irdv to irapaireaov Trj^ i^ov 
 Kai GCpaXeu kcpepero Kara twv Kprjinvwv. It has been noticed 
 before, that nothing but the very dangerous character of 
 the old descent from the Mont Cenis can explain the im- 
 mense losses which the Carthaginians suffered on their 
 way down into Italy, and which amounted, in men alone, to 
 one fourth of the whole number that had gained the 
 summit of the pass. 
 
 (4) It is true that the route of the Mont Cenis is not 
 
 parable aux ruisseaux qui circulent sur le glacier en ete. Plus baut, elle 
 disparaissait completement sous la neige, et je ne yis plus que quelques 
 endroits de son lit, oii la neige etait imbibee. Je rencontrai les der- 
 nieres traces d'eau pres des chalets qui sont situes a Textremite du 
 glacier, et sa presence en cette localite me fit supposer qu'elle prorenait 
 selon toute apparence de quelque source voisine. En revanche, je ne 
 decouvris aucun vestige du torrent de TAar, a Fendroit ou il coulait fete 
 precedent, ni sur aucun point du talus terminal. La cascade du lac de 
 Trubten et le torrent de TOberaar, dont le bruit en ete se fait entendre 
 de loin, avaient aussi completement disparu. On ne reconnaissait leur 
 emplacement qu'a quelques gigantesques gla9ons qui etaient suspendus 
 aux rochers — Agassiz, Etudes sur les Glaciers, p. 373. 
 
152 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 found in the Eoman Itineraries. It will be shewn, how- 
 ever, in the next Chapter, that the route of the Little Mont 
 Cenis is laid down in the ancient map or table of the 
 Koman roads called the Peutingerian or Theodosian Table. 
 Other arguments will also be brought forward there, to 
 shew that this pass was known to the Eomans. 
 
 (5) That the Mont Cenis was not one of the four 
 routes known to Polybius may be strongly questioned. It 
 will be seen, in fact, in the next Chapter, that it has a 
 stronger claim than the Mont Genevre to be considered as 
 the iirepfidcjis Std Tavpivwv, tjv 'Avvi^a^ SifiXOev, mentioned 
 by Strabo on the authority of Polybius. Certainly, neither 
 the Great nor the Little St Bernard can have any right to 
 this designation. If Strabo has cited Polybius' words cor- 
 rectly, and Polybius was not mistaken as to the pass which 
 Hannibal crossed, (and there is no reason for doubting 
 either of these facts ;) then the question, as to the locality 
 of Hannibal's passage of the Alps, resolves itself into the 
 consideration of the respective probabilities in favour of 
 the Mont Genevre and the Mont Cenis, for no other pass 
 leading through the country of the Taurini, such as the Col 
 cle TArgentiere or the Col de la Croix, can have much 
 likelihood on its side ^ 
 
 (6) It may be conceded, with respect to the last objec- 
 tion, that it would be quite impossible now to find any old 
 snow on the descent from the Mont Cenis in the month 
 
 This passage in Strabo has necessarily embarrassed the advocates 
 for the Little St Bernard. The principal of them, M. Deluc, in his dis- 
 cussion of the passage in question, (p. 19) evades the difficulty by taking 
 no notice whatever of Strabo*s assertion (derived from Polybius), that 
 Hannibal crossed the Alps by the road through the country of the Taurini. 
 The author of A Dissertation on the Passage of Hannibal, by a member of 
 the University of Oxford, attempts to overcome the difficulty in the fol- 
 lowing manner : * It seems but fair to state, with regard to this passage 
 m Strabo, that it contains a positive assertion, that Hannibal took the 
 second road, .Ira r^u hh Tavplv^v, f,v 'Auvl^a, 8cij\eeu, and therefore has 
 been brought forward against the hypothesis here supported. But if we 
 prove that Polybius takes Hannibal by another pass, it will follow, that 
 these words are to be ascribed to Strabo, and not to Polybius. Strabo, 
 who wrote in the time of Livy, would probably adopt the opinion of 
 that historian. — p. 19. * 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 353 
 
 of October. Yet it cannot be concluded from this, that it 
 would have been impossible for Hannibal to have found 
 any, for the greater severity of the climate at that time 
 might have caused old snow to be found far below the 
 level within which it is now confined. The statement of 
 Polybius, Twu ydp "AXirewu to. fxev uKpa Ka\ rd irpo^ ras 
 v7r€pfioXd9 dvJKovTay reXem dSevSpa Kal \}^iXd iravr earl, 
 Sid Tocrvvexws eTrineveiv ti^v ^£oVa Kal Oepovs Kat x^t^wi^os* 
 I'd S' iiro ncGtiv Trjv iraptipeiav e^ dfxcpolv toIv fiepolv, vXo- 
 (p6pa Kal SevSpo(j)6pa, Kal to oXov oUiicTifxd earw, — this 
 statement is not true when applied to the present state 
 of the principal Alpine passes. No such barrenness is now 
 observed, and the encroachments of the snow are far less 
 than is here represented. In Polybius' time the snow must 
 have reached much lower than it does now. 
 
 In a previously mentioned note upon this subject in 
 Dr Arnold's History of Borne, a reference is given to 
 Evelyn's Memoirs, by which large tracts of snow are satis- 
 factorily shewn to have existed two hundred years ago, 
 in places where there is now none whatever. Evelyn 
 appears to have crossed the Simplon pass in the first week 
 in September 1646^ ; a time of year when there is perhaps 
 less snow upon the mountains than at any other, for during 
 the month of September the snow recedes little, if at all. 
 The height of the Simplon pass above the sea-level is 6580 
 English feet, about 200 below the Great, and 240 below 
 the Little Mont Cenis. The whole of the Simplon road, 
 in the beginning of September, and indeed during the 
 summer months, is now entirely free from snow, the line of 
 
 1 Evelyn merely says that it was in September, but it appears to have 
 been in the first week in that month. On the day he crossed the Sim- 
 plon he reached Brieg ; thence to Sion, one day ; to Beveretta (Boveret), 
 one day; to Geneva, one day; at Geneva, six weeks; (if exact, forty- 
 two days) to Lyons, two days; to Farrara (Tarare) one day; to Rohan 
 (Roanne) one day; to Orleans, two days; at Orleans, one day; to Paris 
 one day : total, fifty-three days. Evelyn would thus reach Paris the 
 fifty-fourth day after crossing the Simplon. But it was still October 
 when he arrived in Paris ; so that it would bo in the first week in Sep- 
 tember that he crossed the Simplon. 
 
 J 
 
154i 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 perpetual snow being about 2000 feet above the summit 
 of the pass. 
 
 Speaking of the house, where he passed the night, in 
 the hamlet of Simplon^ 4850 English feet above the sea- 
 level, Evelyn says : ' The house was now, in September, 
 half covered with snow, nor is there a tree or bush growing 
 within many miles.* Yet there are, at the present day, 
 numbers of pines about the village of Simplon, and no 
 snow exists in the neighbourhood in the beginning of Sep- 
 tember. 
 
 After getting into a dispute with the inhabitants of the 
 village about a goat, Evelyn's party resumed their journey 
 on the next morning. Alluding to the circumstances 
 under which they left the village, he proceeds : 
 
 'This was cold entertainment, but our journey after 
 was colder, the rest of the way having hen as they told us 
 covered with snow since the Creation ; no man remember d it 
 to he without ; and because by the frequent snowing the 
 tracts are continualy fill'd up, we passe by severall tall 
 masts set up to guide travellers, so as for many miles they 
 stand in ken of one another like to our beacons. In some 
 places where there is a cleft between 2 mountaines the 
 snow fills it up, whilst the bottome being thaw'd leaves as it 
 were a frozen arch of snow, and that so hard as to beare 
 the greatest weight ; for as it snows often, so it perpetualy 
 freezes, of which I was so sensible that it flaw'd y® very 
 skin of my face. 
 
 ' Beginning now to descend a little, Capt. Wray's horse 
 (that was our sumpter and carried all our baggage) plung- 
 ing thro' a bank of loose snow slid downe a frightfull pre- 
 cipice, which so incens'd the choleriq cavalier his master 
 that he was sending a brace of bullets into the poore beast, 
 least our guide should recover him and run away with his 
 burden ; but just as he was lifting up his carbine we gave 
 such a shout, and so pelted y^ horse with snow-balls, as 
 with all his might plunging through the snow he fell from 
 
 1 Although Evelyn mentions this place as being on the summit of the 
 mountain, it is plain that he is speaking of the present Tillage of Sim- 
 plon, and that the highest point of the pass was further on. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 155 
 
 another steepe place into another bottome near a path we 
 were to passe. It was yet a good while ere we got to him, 
 but at last we recover'd the place, and easing him of his 
 charge hal'd him out of the snow, where he had been 
 certainly frozen in if we had not prevented it before night. 
 It was as we judy'd almost two miles that he had slid and 
 falVn, yet without any other harme than the benuming of 
 his limbs for y^ present, but with lusty rubbing and chafing 
 he began to move, and after a little walking performed his 
 journey well enough. All this way, afifrited with the dis- 
 aster of this horse, we trudg'd on foote driving our mules 
 before us ; sometimes we fell, sometimes we slid through this 
 ocean of snow, which after October is impassable. Towards 
 night we came into a larger way, thro' vast woods of pines 
 which clothe the middle parts of these rocks ^* 
 
 It is sufficiently clear, from this account of Evelyn's, 
 compared with the present state of the Simplon, that the 
 snow has lost an immense quantity of ground in the Alps 
 within the last two centuries : and it cannot be concluded, 
 especially when the statement of Polybius as to the general 
 character of the Alps and the Alpine passes is taken into 
 the account, that it would have been impossible, because 
 the descent from the Mont Cenis is now in summer 
 entirely free from snow, that Hannibal should have found 
 permanent snow in that place when he descended into 
 Italy. 
 
 Having now answered the objections of M. Deluc 
 against the Mont Cenis, it will be necessary to notice one 
 from another source. This objection is brought forward in 
 the work, entitled A Dissertation on the Passage of Hannibal 
 over the Alps, by a member of the University of Oxford ; a 
 book written, like M. Deluc's, in advocacy of the Little 
 St Bernard. In p. 52, the author observes : 
 
 * The great size of the army, drawing as it evidently 
 did its resources from the country through which it passed, 
 sufficiently proves, that this country must have been well 
 cultivated, and consequently full of inhabitants; and this 
 
 1 Evelyn's Memoirs, Vol. i. pp. 220, 221. London, 1819. 
 
156 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 latter circumstance is confirmed by the host of barbarians, 
 who hung upon his rear, and attacked him on his passage 
 from Scez». No other of the known passages of the Alps can 
 be at all compared with this one (i. e. the Little St Bernard) 
 in these two essential qualities ; and it is only necessary to 
 have passed through the Maurienne or Mont Cenis road, 
 and the road from the Mont Genevre, along the Romanche 
 to Grenoble, to be convinced, that a large army without 
 magazines must have been starved in any attempt to get 
 through them. It is upon this fact, tvhich has never been 
 sufficiently/ considered before, that I am inclined to lay tlie 
 greatest stress, and when united with the agreement of 
 distances, it becomes almost irresistible.' 
 
 It will not be necessary to examine whether this ob- 
 jection is pertinent or not : for, at least as far as regards 
 the Mont Cenis, it has no foundation. The question is 
 merely one of statistics, and may therefore be exactly 
 determined without difficulty. 
 
 The road of the Little St Bernard traverses the pro- 
 vinces of Tarentaise and of Aosta. It enters the province 
 of Tarentaise near a place called Cevins, about half-way 
 between Albertville (L'Hopital Conflans) and Moutiers : it 
 leaves the province of Aosta at the bridge of St Martin, 
 about eleven miles above Ivrea. 
 
 The road, from La Rochette, over the Little Mont 
 Cenis, traverses the provinces of Maurienne and of Susa. 
 It enters the province of Maurienne about two miles from 
 La Rochette: it leaves the province of Susa between 
 Buttigliera and Rosta, about four miles below Avigliana. 
 
 The length of the road through the Tarentaise and the 
 province of Aosta is nearly the same as that of the road 
 through the Maurienne and the province of Susa, the 
 latter being only about six or eight miles longer. The 
 question therefore rests upon the comparative density of 
 population, in the provinces of Tarentaise and Aosta, and 
 
 1 Had this attack happened on the pass of the Little St Bernard, as is 
 here taken for granted, it would not have been at Scez, but at a great 
 distance below, that it was made. This mistake, as previously shewn 
 when discussing the narrative of Polybius, has arisen from a mis-inter- 
 pretation of that author's words. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 157 
 
 in the Maurienne and the province of Susa. This, how- 
 ever, may be previously observed in favour of the Mont 
 Cenis: the mountain route is much shorter by this pass 
 than by the pass of the Little St Bernard, the commence- 
 ment of the Maurienne being about twenty-five miles 
 below the commencement of the Tarentaise. Hannibal 
 would thus have saved about two days of mountain marching 
 by taking the route of the Little Mont Cenis, in preference 
 to that of the Little St Bernard ; for the frontier of the 
 province of Susa is beyond Avigliana, the commencement 
 of the plains, while the frontier of the province of Aosta 
 is about eleven miles above the similar position of Ivrea. 
 
 The statistical table of the population of the Sardinian 
 States on the mainland of Italy, in the year 1846, gives the 
 number of inhabitants to a square kilometre in every pro- 
 vince. It appears from this table, that, of all the thirty- 
 nine Sardinian provinces on the mainland of Italy, that of 
 Aosta is the most thinly populated. Next in order comes 
 the province of Ossola ; and the third is the Tarentaise \ 
 The Maurienne certainly comes fourth, and is, no doubt,* 
 thinly populated. It surpasses, however, by one-sixth^ 
 in density of population, the province of Tarentaise ; while 
 the population of the province of Susa is more than twice 
 as great, in point of density, as that of the Tarentaise, 
 and, a fortiori, as that of the province of Aosta. 
 
 The following are the numbers of inhabitants to a 
 square kilometre, given by the statistical table : 
 Pass of Little St Bernard. 
 
 Province of Aosta 24.45 
 
 Province of Tarentaise . . . , ' . 25.83 
 Pass of Mont Cenis. 
 
 Province of Maurienne . . , , 30.16 
 Province of Susa 55^91 
 
 It appears then, that the provinces through which the 
 Mont Cenis road passes, support a much thicker population 
 than the corresponding provinces traversed by the road of 
 the Little St Bernard. An argument, however, can hardly 
 
 1 The Tarentaise can thus hardly be called * a very populous country.' 
 See Dissertation on the Passage of Hannibal, p. xvii. 
 
158 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 be drawn in favour of the Mont Cenis from this circum- 
 stance : for Hannibal, beyond all doubt, followed a known 
 path across the Alps, without being influenced by any con- 
 siderations of population or fertility. Besides, the Car- 
 thaginians carried some provisions with them, and are also 
 said to have suffered from the failure of their supplies. 
 They clearly did not depend entirely for provisions upon 
 the productions of the districts through which they passed. 
 
CHAPTER XI. 
 
 Antiquity of the Pass of the Mont Cenis.—First mentioned by name in 
 the year 755 A.D.-Probably known in the sixth century—Described 
 by Ammianus Marcellinus in the fourth century.— The road over 
 the Little Mont Cenis laid down in the Peutingerian table.— The 
 Mont Cenis crossed by Julius Cssar.-Probably mentioned by 
 btrabo as M« pass through the country of the Taurini, which Hannibal 
 crossed. — Conclusion. 
 
 T i^ P''<*PO«ed in this Chapter to consider the antiquity 
 J. of the route of the Mont Cenis, a pass of which it 
 has been generally asserted that the Romans were igno- 
 rant. Yet this way across the Alps seems always to have 
 been well known. For the last thousand years and up- 
 wards it has, beyond all dispute, been in constant use. It 
 IS mentioned by name for the first time in the Continua- 
 tion of the Chronicle of Fredegar, where it is said to have 
 been crossed by Pepin in the year 755 : • Rex Pippinus 
 cum exercitu suo Monte Cinisio transacto. &c.' Before 
 these times it has been supposed that the pass was un- 
 known : mdeed it might be imagined, from the terms in 
 which it has been spoken of, that the diflSculties which it 
 opposed to any attempt at a passage were of such a 
 formidable nature as would have baffled all the efforts of 
 the Romans; and that they only yielded, at length, for the 
 farst time, before the superior engineering skill of the 
 dark ages*. 
 
 A passage certainly existed in the sixth century be- 
 tween the Maurienne and Susa; for, when the bishopric of 
 Maunenne was re-erected by king Gontran of Burgundy, 
 
 I n nous reste a parler du Mont Cenis. Ce passage ne se tronra 
 
 «^!, vnir '"'."'"^"'^^ ■•<»"^'"«. et il ne paroit pas qu'il ait ete jamais 
 une vo.e romame, ou qu'il ait ete m^me connu des Remains, jdoffroit 
 
 2 Ih h , n *'f '■.'° ''■«'"«' '^""^ '^ ""' "f- '« <=hemin par leJueUn 
 descend de la Grande Croix au Tillage de La Femere.-Deluc. Histoir, 
 du Passage des Alpes par Annibal, p. 23. 
 
>5I 
 
 
 ^\ 
 
 v« 
 
 d 
 
 be 
 
 ^' 
 
 %? 
 
 S 
 
 i. 
 
 n\ 
 
 r 
 
 :^ 
 
 S 
 
 ^r 
 
 % 
 
 v> 
 
 
 1> 
 
 f-H 
 e2 
 
 
 
 -5 t 
 
 PL, 
 
 
 CHAPTER XI. 
 
 Antiquity of the Pass of tlie Mont Cenis.-First mentioned by name in 
 tho year 755 A.D.-Probably known in tlie sixth century.-Dcscribed 
 by Ammianus Marcellinus in the fourth century.— The road oyer 
 the Little Mont Cenis laid down in the Peutingerian table— Tho 
 Mont Cenis crossed by Julius Casar.-Probably mentioned by 
 fetrabo as <Ae pass through the cmcntry of the Taurini, which Hannibal 
 cro3«e(i.'— Conclusion. 
 
 TT IS proposed in this Chapter to consider the antiquity 
 Jl of the route of the Mont Cenis, a pass of which it 
 has been generally asserted that the Romans were igno- 
 rant. Yet this way across the Alps seems always to have 
 been well known. For the last thousand years and up- 
 wards, It has, beyond all dispute, been in constant use. It 
 IS mentioned by name for the first time in the Continua- 
 tion of the Chronicle of Fredegar, where it is said to have 
 been crossed by Pepin in the year 755 : ' Rex Pippinus 
 cum exercitu suo Monte Cinisio transact©, &c.' Before 
 these times it has been supposed that the pass was un- 
 known .-indeed it might be imagined, from the terms in 
 which it has been spoken of, that the difficulties which it 
 opposed to any attempt at a passage were of such a 
 tormidable nature as would have baffled all the efforts of 
 the Romans; and that they only yielded, at length, for the 
 hrst time, before the superior engineering skill of the 
 dark ages'. 
 
 A passage certainly existed in the sixth century be- 
 tween the Maurienne and Susa ; for, when the bishopric of 
 Maunenne was re-erected by king Gontran of Burgundy, 
 
 » n nous reste a parlor du Mont Cenis. Ce passage ne se trouve 
 pomt dans les itineraires remains, et il ne paroit pas qu'il ait ete jamais 
 uno voie -mame ou qu'il ait ete memo connu des Remains. IlofZ 
 
 a pic, et il a fallu ta.Iler en zigzags dans le roc vif, le chemin par lequel on 
 descend de la Grande Croix au village de La Femero.-Deluc, HistZ 
 du Passage des Alpes par Annihal, p. 23. 
 
160 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 (561 — 593), the recently acquired city of Susa and its 
 territory were included in the new diocese ^ The Mauri- 
 enne had previously been comprised in the diocese of 
 Turin 2, and not in that of Embrun, Grenoble, or Taren- 
 taise, a fact which seems also to shew that there was a 
 communication between the Maurienne and Italy. The 
 pass which served as this line of communication would, 
 in all probability, have been that of the Mont Cenis. 
 
 Yet it was not only in these times, but also in those 
 of the Romans, that the pass of the Mont Cenis was known. 
 To make this fact appear, it will be attempted to prove 
 three things: 
 
 (1) The pass of the Mont Cenis is alluded to by Am- 
 mianus Marcellinus, a writer of the fourth century. 
 
 (2) The road over the Little Mont Cenis is laid down 
 in the Peutingerian Tabled and was therefore a Roman 
 way. 
 
 (3) The Mont Cenis was crossed by Julius Caesar, when 
 on his way to intercept the Helvetii in Transalpine Gaul. 
 
 The passage in Ammianus Marcellinus, where the 
 Mont Cenis appears to be referred to, is contained in the 
 
 1 See the document relating to the foundation of this bishopric, 
 quoted below, p. 172. 
 
 2 Gregory of Tours * de gloritl martyrum.' Lib. i. cap. 14. 
 
 3 The Peutingerian Table is an ancient Koman map, found in a 
 library at Speyer in the fifteenth century. It is supposed to have been 
 made originally about the middle or end of the third century of the 
 Christian era, but to have been slightly altered by interpolations in 
 later times. Its object was to lay down the Roman roads, and to give 
 the names of the several stations, and the distances between them. All 
 geographical accuracy is disregarded. The seas are represented as long 
 straits or inlets, and no account is taken of the proper figure of the land, 
 which generally runs in stripes or bands from west to east. Indeed, in 
 the proper sense, the Table can hardly be called a map. Its whole extent, 
 twenty-one feet in length, and one foot in width, is divided into eight 
 sheets. A ninth sheet, which would have completed it on the west, is 
 supposed to have been lost. These divisions are very abrupt, and 
 many words are cut in two by them. One such division will be observed 
 in the annexed copy of a part of the Table, which contains portions of 
 two of the sheets. It is chiefly from the Peutingerian Table and the 
 Antonine Itinerary that our information on the subject of the Roman 
 roads is derived. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 161 
 
 tenth chapter of the fifteenth book of his history. It runs 
 thus : 
 
 ; In his Alpibus Cottiis, quarum initium a Segusione est 
 oppido, pr(^celsum erigitur jugum, nulli fere sine discrimim 
 penetrahile. Est enim e Galliis venientihus prona humilitate 
 deveocum, pendentium saxorum altrinsecus visu terriUle, prw- 
 sertim verno tempore: cum liquente gelu, nivihusque 'solutis 
 flatu calidiore ventorum, per diruptas utrinque angustias, 
 et lacunas pruinarum congerie latebrosas, descendentes 
 cunctantibus plantis homines et jumenta procidunt et 
 carpenta ; idque remedium ad arcendum exitium repertum 
 est solum, quod pleraque vehicula vastis funibus illigata, 
 pone cohibente virorum vel boum nisu valido, vix gressu 
 reptante paullo tutius devolvuntur. Et hsec, ut diximus, 
 anni verno contingunt. Hieme vero humus crustata 
 frigoribus, et tanquam levigata, ideoque labilis, incessum 
 praecipitantem impellit, et patulse valles per spatia plana 
 glacie perfidae vorant nonnunquam transeuntes. Ob quse 
 locorum callidi, eminentes ligneos stilos per cautiora loca 
 defigunt, ut eorum series viatorem ducat innoxium: qui 
 SI nivibus operti latuerint, montanis defluentibus rivis eversi 
 agrestibus praeviis difficile pervaduntur. A summitate autem 
 hujus Italici clivi, planities adusque stationem nomine Martis 
 per septem eoctenditur millia : et hinc alia celsitudo erectior, 
 (Bgreque superabilis, ad Matronce porrigitur verticem, cujus vo- 
 cabulum casus foeminse nobilis dedit. Unde declive quidem 
 iter, sed expeditius adusque castellum Virgantiam patet.' 
 
 The ' Mons Matrona ' is the Mont Genevre ^ ; the * prse- 
 celsum jugum,' in the first sentence, is distinguished from 
 the Mons Matrona, as being situated between Ad Martis 
 and Susa, and thus lying below, and not above, the former 
 station. Its base, that is to say, lay between Susa and 
 Ad Martis ; for no mountain is crossed, but only skirted, 
 on the road between these two places. Ammianus Mar- 
 cellinus, however, is not so clear here as might be wished. 
 He seems not to have fully comprehended the accounts 
 from which he compiled his description : the phenomena of 
 this ' prsBcelsum jugum,' the avalanches, snow-drifts, &c.. 
 
 * See the Itinerary from Bordeaux to Jerusalem. 
 
 11 
 
162 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 are rather confusedly put together ; although the circum- 
 stantial manner in which they are related, and the truth 
 and accuracy of the details given, shew that the autho- 
 rities which he consulted were perfectly trustworthy. It 
 is the expression, * a summitate hujus Italici clivi ' which 
 introduces some difficulty. Yet the point indicated by 
 this word ' summitas ' is easily found. For about ten 
 Boman miles from Susa, in the direction of the Genevre, 
 the Dora flows through a continuous defile, which never 
 opens out into a proper valley till about two miles above 
 Exilles. Here the mountains fall back, and a level plain 
 is entered, which extends, for about six Eoman miles, as 
 far as Oulx, the ancient * Ad Martis.' This level tract is 
 clearly the * planities ' which * adusque stationem nomine 
 Martis per septem extenditur millia^' The 'summitas 
 hujus Italici clivi ' is therefore the end of the defile, about 
 two miles above Exilles, and one below Salbertrand. 
 
 Yet this * summitas ' is clearly not the summit of the 
 'praecelsum jugum,' even if Ammianus supposed it to be 
 so. For it is merely a point in the valley of the Dora, 
 and not the crest of a pass at all. Neither would the road 
 from this point to Susa at all answer to the description of 
 the descent from the * jugum.'' Some other meaning must 
 evidently be found for the * summitas hujus Italici clivi.' 
 
 The employment of the words ' Italicus ' and ' clivus ' 
 leads us readily to the meaning to be affixed to this expres- 
 sion. For all the main chain of the Alps separating Gaul 
 from Italy, and every one of the different mountain-masses 
 into which that chain may be divided, has necessarily an 
 Italian and a Gallic slope, an ' Italicus ' and a * Gallicus 
 clivus.' Now, between the point indicated by Ammianus 
 as the ' summitas,' and the city of Susa, there is a conti- 
 nuous defile, extending from the end of the 'planities' 
 to the beginning of the open valley called the Combe of 
 Susa. This defile is formed on the north by a mountain- 
 mass, which may be considered to begin at Susa, and 
 to end at the * summitas.' The lower extremity of the 
 Italian declivity of this mountain-mass, or mountain, is 
 
 1 Compare Cajsar, Bell Gall. Lib. in. (beginning) where the 'plani- 
 ties* of Octodurus, or Martigny, is mentioned. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 163 
 
 thus at Susa ; the ' summitas hujus Italici clivi,' (i. e.) the 
 upper extremity of the Italian declivity of this mountain, 
 is at the beginning of the • planities,' two miles above Ex- 
 illes. These were the limits of the Italian base of the 
 'praecelsum jugum,' across which lay the passage from 
 Gaul of which we are now in search. 
 
 But over this mountain-mass only three passes exist, 
 regarding the Great and Little Mont Cenis as one. These 
 passes are : the Mont Cenis, the Col de Clairee, and the 
 Col d'Ambin or Col de Galumbre. Yet these two last, of 
 which the names are scarcely known, are merely diflScult 
 mountain-tracks : no * carpentum,' however small and rude 
 a vehicle be supposed to be designated by that term, 
 could ever have crossed either of these cols. Besides, the 
 path from Savoy to the Col de Clairee passes on its way 
 within a few yards of the Col of the Little Mont Cenis, so 
 that the subsequent ascent would be needless. Indeed, the 
 Mont Cenis is the only one of these three ways that can 
 properly be called a pass. The others could never have 
 been used as lines of communication between the valleys 
 of the Arc and the Dora. 
 
 Yet it is not in position only that this * praecelsum 
 jugum' is found to coincide with the Mont Cenis. The 
 characters of the two passes, as far as the descent into 
 Italy is concerned, are also identical. For the descent from 
 Gaul by the 'prsecelsum jugum' was excessively steep, 
 *prona humilitate devexum,' flanked on both sides by 
 frightful precipices, ' pendentium saxorum altrinsecus visa 
 terribile,' and also liable in spring, the time of avalanches, 
 'nivibus solutis flatu calidiore ventorum,' to be swept by 
 these falling masses of snow. The first part of the de- 
 scent from La Grande Croix on the Mont Cenis, by the 
 plain of St Nicholas to La Ferriere, answers precisely 
 to this description. Of its extreme steepness enough has 
 already been said: the danger from avalanches, and the 
 precipices on both sides of the way, are also noticed in 
 several descriptions of the pass. 
 
 In a small work, entitled Guide du voyageur a Suse et 
 au passage du Grand Mont-Cenist (Susa, 1830), the latter 
 part of the ascent from Susa is thus described ; 
 
 11—2 
 
164 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 *Au dessous de la route actuelle^ et a sa droite, un 
 
 peu avant que d'arriver a la plaine de Saint Nicolas, se 
 
 trouve une ancienne galerie en magonnerie^, que Ton 
 
 appelait la route couverte; elle avait ete eonstruite pour 
 
 garantir les voyageurs de Pimpetuosite des vents et de la 
 
 coulee des neiges ; elle a 140 metres de longueur et 4 metres 
 
 80 centimetres de largeur. Get ouvrage est aujourd'hui 
 
 presqu'entierement enseveli sous les deblais que Ton a 
 
 dA faire pour former la nouvelle route qui rejoint Panci- 
 
 enne a Tentree de la plaine de St Nicolas, dont I'horrible 
 
 beaute penetre le voyageur d'admiration et d'effroi. Ce 
 
 petit bassin est domine du cote droit par des rochers tres- 
 
 escarpes qui semblent menacer les airs. A sa gauche se 
 
 trouve la nouvelle route qu'on avait premierement essaye 
 
 d'ouvrir a travers d'un rocher taille a pic et d'une hauteur 
 
 prodigieuse^ mais que, malgre les grands travaux qu'on 
 
 avait deja faits, Ton a ete force d'abandonner, a cause de la 
 
 chilte continuelle des masses enormes de rocher qui se 
 
 detachaient de la montagne, et des dangers imminents des 
 
 avalanches auxquels on etait frequemment expose.' 
 
 M. Albanis de Beaumont, {Les Alpes Grecques et Cottt- 
 enneSf Vol. ii. p. 652) speaking of the descent from the 
 Mont Cenis, says of the plain of St Nicholas : 
 
 ' JEUe est couverte d'une belle verdure et entouree de 
 rochirs abruptes, dont les sommets se perdent dans les 
 nues : c'est du haut de ces memes rochers que se pre- 
 cipitent en cascades les eaux limpides de la Cenise. Apres 
 avoir traverse cette plaine, Pon passe sous une longue 
 voMe* que Tancien gouvernement Sarde avait fait con- 
 struire afin de prevenir les accidents autrefois si communs sur 
 ce passage, occasiones par les avalanges toujours si dange- 
 reuses et si fatales lors de la fonte des neiges ; ici ces ava- 
 langes sont si considerables, qu'en plusieurs occasions elles 
 
 1 The present carriage-road is here spoken of. 
 
 2 This was on the old road from La Ferriere. 
 
 3 The *promontorium' from which Italy is seen. The cliff in which 
 it terminates is pierced by a timnel, through which the carriage-road at 
 first ran. 
 
 4 The new road was at this time (1802) not yet made. The * voute' 
 is the * route couverte ' previously mentioned. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. IQ^ 
 
 mt}T}^^i t^' ^^"^^^^^^^«^ent de rotation sept a huit 
 mulets a la fois amsi que leurs conducteurs, qu'elles ont 
 ensuite ensevehs dans le precipice affreux ou coulent les 
 eaux de a Cenise, et d'oii leurs cadavres n'ont ete retires 
 qu apres Pentiere fonte des neiges/ &c. 
 
 The descent from the Mont Cenis into Italy is thus 
 perfect y identical in character with the descent from the 
 praecelsum jugum ' of Ammianus Marcellinus, with which 
 It has also been concluded to coincide, in respect of 
 position. It may, therefore, be fairly supposed, that the 
 Mont Cenis was, m fact, the pass between Gaul and Italy, 
 ot which Ammianus speaks in this place. 
 
 Before proceeding with the question of the Eoman 
 road laid down in the Peutingerian table, it will be neces- 
 sary to determine the name of the tribe which dwelt in the 
 Upper Maurienne. This tribe will be found to be the 
 Cottian, as distinguished from the Vocontian, Caturiges 
 
 It is generally considered that there was in the Alps 
 but one tribe bearing the name of Caturiges, and that 
 they dwelt about the town of Caturig^ or Caturiges, now 
 Chorges, between Gap and Embrun. There is, however 
 no reason why there should not have been two tribes or 
 two branches of the same tribe, identical in name, 'in- 
 stances of such a separation occur elsewhere. In the 
 valley of Susa dwelt the tribe of the Segusini or Segusiani • 
 and m the modern Lyonnais there was also a tribe called 
 Segusiam. The Nantuates are mentioned by Csesar^ as 
 dwelling on the Upper Rhine ; but there was also a tribe 
 called Nantuates to the S.E. of the lake of Geneva, and 
 perhaps another people of the same name near the modern 
 Nantua m the department of the Ain. Pliny, also ex- 
 pressly speaks of people descended from the Caturiges 
 though of different name ; a fact which seems to intimate 
 that they were not confined to one spot in the Alps, but 
 might have been scattered in several places^ 
 
 , 1 Bell Gall Lib. iv. 
 
 2 Ligunim celeberrimi ultra Alpes Sallyi, Deceates, Oxubii : citra 
 Veneni^ et Caturigibus orti Vagienni, Statyelli, Vibelli, Magelli, Eubu. 
 nates, Casmonates, VeUates, et quorum oppida in era proxima dicemus -^ 
 
166 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 ' I 
 
 That there was a people in the Alps called Caturiges, 
 who dwelt in some other district than the neighbourhood 
 of Chorges, seems certain. Strabo, in his fourth book, 
 speaks of the Centrones, the Catoriges, the Veragri, and 
 the Nantuates, as dwelling on the mountains above the 
 Salassi. Now the limit of the Salassi on the south seems 
 to have been the Stura, or the Dora Susina^ : they pro- 
 bably inhabited all the valleys on the Italian side of the 
 main chain of the Alps, from the Eoche Melon to Monte 
 Eosa. Yet no part of their territory could be near Chorges, 
 or its district. The whole breadth of the Cottian land 
 intervenes between Embrun, the frontier town of the 
 people of Chorges, and the Eoche Melon, the nearest 
 point of the country of the Salassi. The peoijle of Chorges, 
 also, could not possibly be associated, in respect of situation, 
 with any one of the tribes of the Centrones, Veragri, or 
 Nantuates, or be spoken of as a Cottian tribe. For the 
 boundary of the Cottian territory was at Embrun, accord- 
 ing to Strabo. Chorges would thus, on the same authority, 
 have belonged to the Vocontii^. 
 
 The Caturiges, or Catoriges, who bordered on the 
 Salassi, must have inhabited the Upper Maurienne. For 
 the other three tribes bordering on the Salassi, the Ver- 
 agri, the Nantuates, and the Centrones, occupied the 
 Lower Vallais, and Eastern Savoy as far (inclusive) as the 
 Tarentaise. The Caturiges would therefore be sought, 
 either in the Upper Vallais, or else in the Upper Mau- 
 rienne or the valley of Susa. But they could not have 
 dwelt in the Upper Vallais, for they are never mentioned 
 among the people inhabiting that district, and were, be- 
 sides, one of the Cottian tribes ^ : and the only parts of the 
 
 Nat. Hist. Lib. m. Cap. 5. And again : Sunt prsDterea Latio donati 
 incolse, ut Octodurenses, et finitimi Centrones, Cottianso civitates ; Catu- 
 riges, et ex Caturigibus orti Vagienni, Ligures, et qui Montani vocantur. 
 — Lib. in. Cap. 20. 
 
 1 The existence of a village called * Salassa,' or * Salazza,' on the right 
 bank of the Orca, seems to Intimate that the Salassi extended to the 
 south beyond that river. 
 
 2 Lib. IV. Cap. 1. 
 
 3 The names of the Cottian tribes are found on the Arch of Susa. 
 The following is the inscription : Imp. Cacsari . Augusto . Divi . F. Pontifici . 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 167 
 
 Cottian territory which touched the country of the Salassi 
 were the Upper Maurienne and the valley of Susa. But 
 the valley of Susa was inhabited by the Segusini: the 
 Caturiges, therefore, should be sought in the Upper Mau- 
 rienne. 
 
 Again, the Upper Maurienne borders on the Tarentaise. 
 It is, accordingly, with the Centrones, the inhabitants of 
 the Tarentaise, that we find the Caturiges leagued, when 
 Ca3sar (Bell. Gall. Lib. i.) crossed the Alps to intercept the 
 Helvetii, and had to encounter the opposition of the 
 mountaineers. 
 
 In the enumeration of the Cottian tribes on the Arch of 
 Susa, the Caturiges are preceded by the Segusini and the 
 Belaci, and followed by the Medulli. Now the Segusini 
 inhabited the valley of Susa ; and the Belaci are placed in 
 the valley of Bardoneche, where the village of Beaulard is 
 supposed to preserve their name. The Medulli dwelt in 
 the Lower Maurienne. AVe thus find the Caturiges either 
 certainly, or probably, connected in point of position 
 with five tribes ; the Salassi, the Centrones, the Medulli, 
 the Belaci, and the Segusini. But the districts occupied 
 by these fi\e tribes completely encircle the Upper Mau- 
 rienne. It is consequently, on the whole, highly probable, 
 not to say certain, that the Cottian Caturiges inhabited 
 the Upper Maurienne. 
 
 The question of the former existence of a Roman road 
 over the Little Mont Cenis may now be considered. In 
 the part of the Peutingerian table accompanying this book, 
 four Roman roads across the Alps will be found, three 
 proceeding from Turin and one from Ivrea ; the first three 
 being coincident for a certain distance. The first of these 
 roads may be called the Turin and Aries road. It leads 
 
 maxumo . tribunic . potestate . xv . imp . xin. M. Julius . Regis . Donni . 
 F. Cottius . Prsefectus . ceivitatium . quse . subscriptae . sunt . Segoviorum. 
 Segusinorum . Belacorum . Caturigum . Medullorum . Tebaviorum . Ada- 
 natiura . Savincatium . Egidiniorum . Veaminiorum . Venicamorum . Ime- 
 riorum . Vesubianorum . Quadiatium . et ceivitates qure sub eo prsefecto 
 fuerunt. 
 
■iiiwi wyi i ii iii 
 
 168 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 from Turin to Aries, by Susa, the Mont Genevre, (marked 
 Alpis Cottia on the Table) Brian^on, Embrun, Gap, Sis- 
 teron, and Apt. This road is, as appears by comparison 
 with the Itineraries, correctly laid down. The second road 
 may be called the Turin and Valence road. It is repre- 
 sented, (erroneously, as will be shewn) as branching off 
 from the first road on the Mont Genevre. The third road 
 may be called the Turin and Vienne road. It is laid down, 
 (erroneously also) as striking off from the second road close 
 to the summit of the Mont Genevre (no station intervening). 
 The fourth road may be called the Ivrea and Vienne road. 
 It leads by the valley of Aosta, the Little St Bernard, the 
 valley of the Isere, Lemincum (near Chambery), the Char- 
 treuse mountains, and Bourgoin, to Vienne. It was the 
 third of these roads, the Turin and Vienne road, which 
 passed over the Little Mont Cenis. It has, however, 
 hitherto been considered to have passed over the Col du 
 Lautaret, between Brian9on and Grenoble. 
 
 The second of these roads, the Turin and Valence 
 road, is manifestly laid down incorrectly. The station of 
 Lucus (Luc en Diois) is represented as being only 46 miles 
 from the summit of the Mont Genevre. Yet Luc, even in 
 a straight line, is 70 Eoman miles from that mountain : 
 the road to Luc must, then, clearly branch off from the 
 Turin and Aries road at some nearer point than the Mont 
 Genevre. Besides, there is only one natural way of 
 reaching Luc from the Mont Genevre ; and that is by the 
 valley of the Durance, down which ran the road to Aries. 
 The road to Luc is not likely to have branched off before 
 reaching Gap. Now a very important line of way, that 
 given in the Itinerary from Bordeaux to Jerusalem, did 
 pass through Luc, and fell into the Turin and Aries road 
 at Gap: so that either this road must be the one erro- 
 neously laid down in the Peutingerian table, or else must 
 be totally omitted there, which is not very probable in the 
 case of so important a highway. According to the Itine- 
 rary of Antoninus (Wesseling's Itineraria Romanorum) the 
 distance between Luc and Gap (Vapincum) was 50 miles, 
 a number differing but little from the 46 miles of the 
 Peutingerian table. It thus appears that the Turin and 
 
 HANNIBAL'S PASSAGE OF THE ALPS. ]69 
 
 Valence road of that Table left the Turin and Aries road 
 at Gap, and not on the summit of the Mont Genevre ^ 
 
 The error which has vitiated the Peutingerian table in 
 this place may now be readily perceived. The author has 
 confounded the * Alpis Cottia' with the ' Alpes Cottia ' 
 Por, though the Cottian territory terminated at Embrun 
 the name of the Cottian Alps, when applied to mountains] 
 might have been extended to Gap, and even beyond. The 
 Turin and Valence road did not separate from the Turin 
 and Aries road 'in Alpe Cottia,' but 'in Alpibus Cottiis;' 
 not on the Mont Genevre, but somewhere in the Cottian 
 Alps. The case would be similar with respect to the 
 Turin and Vienne road. 
 
 Having now rectified the course of the Turin and Valence 
 road, the Turin and Aries road may be left out of consi- 
 deration : for it was from the Turin and Valence road that 
 the Turin and Vienne road struck off But at what point 
 did It strike off? This is as yet uncertain: all that we 
 can conclude is, that this point was somewhere in the 
 Cottian Alps. There are consequently four roads, each of 
 which may have a claim to be considered as the Turin and 
 Vienne road : (1) the road of the Croix Haute, which strikes 
 off northward from the present road to Luc, at Aspres • 
 (the Roman road from Gap to Luc seems to have passed 
 a little to the south of Aspres) ; (2) the road from Gap to 
 Grenoble by the valley of the Drac : (3) the road from 
 Briangon to Grenoble by the Col du Lautaret ; and (4) the 
 road over the Mont Cenis. 
 
 Now the Peutingerian table gives 8b Roman miles as 
 the distance from Vienne to the point where the Turin 
 and Vienne road faUs into the Turin and Valence road 
 But, by the nearest of the four roads mentioned, the route 
 of the Croix Haute, the distance from Vienne to Aspres 
 
 ^ The Peutingerian stations between Gap and Luc should probably 
 both be read * Geminaj.' The only station mentioned between these two 
 places m the Itinerary of Antoninus is * Mons Seleucus,' now *La Batie 
 Mont Saleon.' The Itinerary from Bordeaux to Jerusalem gives, from 
 Luc : Mutatio Vologatis, (inde ascenditur Gaura Mons), Mutatio Cam- 
 bono, Mansio Monte Seleuci, Mutatio Dayiano, Mutatio ad fine, Mansio 
 Vapinco. 
 
170 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 171 
 
 /, 
 
 is more than 120 Eoman miles. The Peutingerian table 
 is, therefore, doubly erroneous with respect to the Turin 
 and Vienne road : that road does not branch off from the 
 Turin and Valence road at the place assigned, nor does it 
 reach Vienne at a distance of 85 miles from the point 
 where it strikes off. 
 
 We have, therefore, a second error to remove. The 
 principle to be applied is suggested by the detection of the 
 error on the Turin and Valence road. For the road from 
 Luc did not reach the summit of the Cottian Alp at a 
 distance of 46 miles from Luc, but it fell, at about that 
 distance, (^. e. at Gap) into the road leading to the Cottian 
 Alp. The Turin and Vienne road, as it cannot reach 
 Vienne after a course of 85 miles from the point where it 
 leaves the Turin and Valence road, should thus, it may be 
 expected, fall, at the termination of that length of way, 
 into a road on the Table leading to Vienne. But this road, 
 into which it falls, must be the Ivrea and Vienne road, 
 for there is no other. The Turin and Vienne road must 
 either fall into this, or return into the road from which it 
 diverged ; a supposition which is hardly possible. 
 
 The question, therefore, which has now to be deter- 
 mined, is this : which of the four roads, the route of the 
 Croix Haute, the route of the valley of the Drac, the 
 route of the Lautaret, or the route of the Mont Cenis, (the 
 Little Mont Cenis .will be selected) falls, at a distance of 
 about 85 Eoman miles from the point where it leaves the 
 Turin and Valence road, into the Ivrea and Vienne road. 
 
 85 Roman miles are equivalent, neglecting fractions, to 
 125 French kilometres, or 78 English miles. 
 
 For the three roads, by the Croix Haute, by the valley 
 of the Drac, and by the Lautaret, the nearest station on 
 the Ivrea and Vienne road is that of Augustum, now Aoste 
 de St Genix. From Aspres to Voirons, by the Croix 
 Haute and Grenoble, the distance is given as 124 kilo- 
 metres. From Gap to Voirons, by the valley of the Drac, 
 the distance is 126 kilometres. From Brian9on to Voreppe, 
 by the Lautaret, the distance is 135 kilometres. Voirons 
 is about 20 Roman miles from Aoste de St Genix, and 
 9 kilometres from Voreppe. 
 
 
 By either of the two routes, that of the Croix Haute, 
 or that by the valley of the Drac, the distance to the Ivrea 
 and Vienne road is thus about 20 Roman miles longer that 
 it should be. By the Lautaret, the distance is more than 
 30 Roman miles in excess. 
 
 The distance from Susa to Aiguebelle by the Little 
 Mont Cenis (see page 91) is 82-i. Roman miles. The dis- 
 tance from Aiguebelle to Maltaverne is 4^ Piedmontese, 
 or 7 Roman miles. The 85 Roman miles, given in the 
 Table, will therefore bring the road to a point about 3 
 Roman miles below Aiguebelle, and 4 from Maltaverne ; 
 in fact, exactly to the point where the valley of the Arc 
 joins that of the Isere, down wdiich ran the road from 
 Ivrea to Vienne. 
 
 The route of the Little Mont Cenis seems thus to be 
 the road of which we are in search. The station, at or 
 near which the two roads to Vienne would join, must have 
 been that of Mantala, which lay at an equal distance (16 
 Roman miles) from Lemincum, (Lemenc near ChamberyM 
 and Ad Publicanos, (Le Fay probably 2), which w^as not 
 very far from the modern Albertville. The station of 
 Mantala has been fixed at Freterive, or St Pierre d'Albigny, 
 or Bourg TEvescal, all on the right bank of the Isere' 
 nearly opposite IMaltaverne. There seems, however, not 
 much doubt, from the resemblance of names, that it should 
 rather be placed at Maltaverne, otherwise Malataverne, a 
 word probably compounded of Mantala and Taherna; and 
 that the Roman road down this part of the valley of the 
 Isere ran on the left, and not on the right bank of the 
 river ^ 
 
 The comparison of the names of the stations on the 
 Roman road from Turin to Vienne, with the names of the 
 places with which they are identified on the supposition of 
 that road having crossed the Little Mont Cenis, will, it is 
 
 1 Maltaverae is 11 Piedmontese, or 18 Roman miles, from Chambery. 
 
 2 See Note A. at the end of the book. 
 
 3 Roman remains have been found at Ayton, on the left bank of the 
 Isere, as well as at places on the right bank. See Les Alpes Orecques et 
 Cottimnes, Vol. n. p. 601. 
 
172 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 conceived, remove all doubt as to the fact of the Roman 
 road having passed this way. 
 
 The following table will shew the modern places which 
 correspond to the old Eoman stations, with the actual 
 distances between the various points in Roman miles, and 
 the same distances as given in the Peutingerian Table. 
 
 Ancient Stations. 
 
 Segusio . 
 Stabatio 
 
 Modern places. 
 
 Actual dis- I Peutingerian 
 tanees in distances in 
 
 Roman 
 Miles. 
 
 Susa. 
 
 La Ferriere 
 
 Durotincum Granges de Dervieux... 
 
 Mellosedum 
 Catorissium 
 CuJabo 
 
 Morginnum , 
 Turecionicum 
 Mantala 
 
 89^ 
 
 Bramans 
 
 Villarodin 
 
 Orelle * 
 
 St Jean de Maurienne... 
 
 La Chapelle^ 
 
 Maltaverne 
 
 Total 
 
 Stabatio, the first station on the way from Susa, is here 
 Identified with La Ferriere. This village lies half way up 
 the great ascent of the Mont Cenis, beginning at La No- 
 va ese, and ending at La Grande Croix, and is the only 
 village or hamlet between those two places. Its site would 
 thus naturally be chosen for a halting place on the ascent. 
 ^ The station of Durotincum falls at the Gran(;es de Der^ 
 vieiuc a place on the plateau of the Little Mont Cenis 
 In the word Dervieua^, some vestige of the Roman name 
 probably remains. The termination vieuw ^QQias^ in ad- 
 dition, to indicate the antiquity of the place. 
 
 The word Durotincum is, as it stands, almost pure 
 Celtic. It resolves itself readily into Dur-tin-cum, all which 
 syllables are Celtic words. Dur^' signifies 'water' and 
 may mean here 'the river' or possibly 'the Dora.' ' Tin^ 
 signifies * a beginning,' the ' source ' or ' origin ' of the 
 river. Cum will be the same as the Welch Cwm, ' a valley ' 
 or ' a hollow in the mountains.' In the ' Combe of Susa ' 
 
 1 If XVI. were read instead of XIV., the hamlet of Tardy would 
 correspond to Turecionicum, 
 
 2 O'Reill/s IHsh, Diet, s ibid 
 
 HAN]S[IBAL'S PASSAGE OF THE ALPS. 
 
 173 
 
 and the ' Combe of Malval,' the word is still used in this 
 sense among the Alps. 
 
 ^ Durotincum will thus mean, 'the hollow in the moun- 
 tains where the river (or Dora) rises.' Now it is on the 
 Mont Cenis that the Cenise, one of the affluents of the 
 Dora, takes its rise. The longest of the little rills, which 
 fall into the lake, and give origin to the Cenise, has its 
 source on the Little Mont Cenis. The plateau of the 
 Little Mont Cenis, enclosed by mountains, may therefore 
 properly be called Durotincum. 
 
 The next station bears the name of Mellosedum. This 
 word also is Celtic in origin. The first syllable exists in 
 Melrose ^ Meldrum, &e. It is a variation of the word 
 Maol, signifying * bare,' 'barren/ * bleak V The word 
 Mellosedum seems thus to be MaoUon-sead, Lon signifying 
 'a meadowV and Sead *a way,' * a road,' or 'a seatV 
 The signification of Mellosedum would therefore be, ' the 
 station on the bare (or open) meadow.' 
 
 These * open meadows ' are common in the Alps. Most 
 travellers in those mountains must remember having fre- 
 quently emerged from some ravine or narrow valley into 
 a small open plain or basin, surrounded by a circuit of 
 mountains, and sometimes almost entirely bare of every- 
 thing but grass. These open spaces are called *Matt' 
 in the German, and * Pra ' or * Praz ' in the French or 
 Piedmontese Alps, words of continual recurrence, and 
 all signifying 'ameadow^' Names nearly equivalent in 
 
 1 'Metros ov Maol Ross, (i. e.) the bare promontory.' — Carlisle, Topo* 
 graphical Dictionary of Scotland. 
 
 * Armstrong's Gaelic Diet. 
 
 8 Armstrong's Gaelic Diet. In Anderson's Guide to the Highlands, 
 a *Lonn' is defined as a 'meadow- strath.' 
 
 4 Armstrong's Gaelic Diet. O'Reilly's Irish Diet. 
 
 « In the Diccionario Geograjlco-estadistico-historico de Espana by 
 Madoz, fiye places bearing the name of * Melon' are enumerated; all in 
 the province of Orense in Galicia, a mountainous country whose original 
 inhabitants were of Celtic origin. One of these places is described as 
 being situated in a hollow surrounded by mountains, * en una hondonada 
 circuida por los montes denominados Faro y Carbela,' a locality of the 
 same kind as a 'Matt' or *Pra' of the Alps. The word Melon is also 
 found in the name of the mountain of the Roche Melon, an appella^ 
 
174 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 meaning to Mellosedum are still found. Andermatt (i.e.) 
 {Der On) an der Matt is one of these. Zermatt, probably 
 the same as Zurmatt, (i. e.) zu der Matt, is another. 
 
 The position of Bramans is accurately defined by the 
 word Mellosedum, for it lies on a small open plain, 4000 
 feet above the sea, and is reached by the traveller from 
 Modane after passing through the long defile guarded by 
 the fort of L'Esscillon. It lies also, relatively to the 
 Little Mont Cenis, in the same situation as Andermatt and 
 Zermatt lie, relatively to the passes of the St Gothard and 
 the Cervin ; the valley of the Arc being entered from the 
 Little Mont Cenis at Bramans, and the valleys of the 
 Eeuss and the Visp in the immediate neighbourhood of 
 Andermatt and Zermatt, both these latter places lying on 
 the open ground at the feet of their respective passes. 
 
 The name Bramans seems, also, nearly identical in 
 meaning with Mellosedum, Bramans is Bra-mans. But Bra 
 appears the same word as Pra, the term by which an open 
 tract or Maol-lon is now designated. Mans also is easily 
 corrupted from Mansio, the name which would be given 
 to one of the more important stations on a Roman road^ 
 Brama7is might thus be Prati Mansio, Probably, when 
 the Celtic language was dying out in the Alps, and a Latin 
 or Romance dialect was taking its place, the old name Mel- 
 losedum Ml into disuse, and was replaced by an equivalent 
 term of Latin origin. 
 
 In the next station, Catorissium, (Villarodin), the name 
 of the Caturiges or Catoriges, the people of the Upper 
 Maurienne, appears. All the names of the other stations 
 are in the ablative. Catorissium is clearly the genitive of 
 Catorisses, and vico or some similar word, must be under- 
 stood 2. 
 
 tion which may bo taken as nearly identical in meaning with Matterhorn 
 (Ceryin). The Roche Melon rises above the fields of La Novaleso in 
 the same manner as the Matterhorn above those of Zermatt. 
 1 See the Itinerary from Bordeaux to Jerusalem, passim. 
 In the Peutingerian table, Addeam occurs under the word Catoris- 
 sium, This belongs properly to the word Bocontiorum, which is found 
 below on the road from Gap to Valence ; the station Dea Vocontiorum 
 or Ad deam Vocontiorum being indicated. The name of the station 
 Catorissium may possibly, however, have been, in full, Vea Catorissium, 
 
 v 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 1 75 
 
 In Culabo we have again a Celtic term. It is preserved 
 most nearly in the Gaelic Culaobh^ and appears in the 
 Piedmontese Culaton^ The name of the Col de la Cula a 
 pass at the e.vtremit^j of the valley of Barcelonnette, seems 
 tp derive its designation from this old word, which defines 
 
 its position. ciiiica 
 
 The word Culabo will signify ' extremity ' or ' end •' it 
 corresponds to the Ad fines of the Romans. Orelle, which 
 IS Identified with Culabo, would be probably the last village 
 of the Caturiges ; for it is to the Upper or Caturig^n 
 
 attached 3. The word Orelle seems also to signify ^ex 
 tremity,' being apparently a diminutive of ora It is' 
 most likely connected with the heraldic orle 'a border' 
 and with the ItaHan orlo and the Spanish orilla, both si^nL 
 fying 'extremity.' In Ponza's Piedmontese-ltalian Dic^ 
 tionary, it will be seen that eula-toni^ rendered in Italian 
 orUiccio, clearly indicating the identity of idea in the two 
 cases* 
 
 Culaho like 3Mlosedum, probably lost its original name 
 with the decay of the Celtie language, and became known 
 by the equivalent Romance appellation of Orelle 
 
 The station oi Morginnum falls at St Jean de Mauri- 
 enne. Here the similarity of name is at once apparent- 
 but It ,s m reality still closer than it seems. For The 
 former name of St Jean was simply Maurienne\ and not 
 
 ne word Dm is probably the Celtic Tcagh, ' a house.' See Recherche ,ur 
 le, langues CelUques, par W. F. Edwards, pp. 466, 467 
 
 ^ • ?i!f*' *f **'tP''* of anything.'-Armstrong's Gaelic Dictionary. 
 Culaton estvemit^ parte deretana : culaton cCl pan, orliccio, estre- 
 m.t^ del pano.'-Ponza. Vocaholario Piemontese-Italiano. ' Cular^TZ- 
 mfie proprement _?;«, recule, cnrenm-'-Vm, Hiuoirc de GrenoUe, p. 2 
 
 3 If the frontier of the Caturiges was, in the time of Hannibal, near 
 Ore e, he would, after passing the Rock of Baune, soon enter int the 
 terruory of a different tribe from the one which had attacked him there! 
 That such was the case, has been previously conjectured. See p. 112 
 
 ■• Qucsta piccola citta, di cui s' ignora la fondazione, sino dall' anno 
 3 avea .1 nome di Moriana ed un proprio vesco.o: ne prese i noma 
 di S. Giovanni che dopo la meta del secolo VI, in cui il re Gontranno 
 d, Borgogna, scacciatino i Langobardi, vi fcce fabbricare la cattedralo 
 dedicata al Santo precxirBoro.—Iti,ierario dcgli Stati Sardi. 
 
176 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 St Jean de Maurienne, In mediaeval Latin it is called Maxi- 
 rienna or Morimna^, while the more complete, and probably 
 more ancient, form of the word is Maurigenna or Morigmna\ 
 Little doubt can be entertained of the identity of the Ro- 
 man Morginnum and the Morigenna of the dark ages^ 
 
 The station of Turecionicum falls at La Chapelle, an 
 important village in the Lower Maurienne. It has been 
 already spoken of in the description of the route of the 
 Mont Cenis. (Chap. vi. p. 82.) 
 
 From Mantalay (Maltaverne) the Roman roads to 
 Vienne from the Little Mont Cenis and the Little St 
 
 * See Hadriani Valesii Notitia GalUarum, where several passages are 
 quoted from mediaeval writers, in which St Jean de Maurienne is called 
 * urbs Maurienna' and * civitas Maurienna/ or, in either case, * Morienna.' 
 In Baronius it is mentioned by the name * Geneva Maurienna,' as one of 
 the six sees subject to Vienne. *Porro ilia sex oppida vel civitates, Gra- 
 tianopolis videlicet, Valentia, Dia, Alba, Vivarium, Geneva Maurienna, 
 in ejus tanquam in proprii Metropolitan! obedientia et subjectione por- 
 maneant.' — Baronii Annates, Vol. xviii. Ad. 1120. It seems to have 
 been ceded by the Franks, under the name of Geneva, to Theodoric the 
 Ostrogoth. See Lebeau, Histoire du Bas-Empir£, Vol. vii. p. 196. 
 
 2 Auctoritas quod ex antiquo Morinensis ecclesia Viennensi metropoli 
 suhdita/uit. 
 
 Audiens autem gloriosus Guntramnus rex de 
 
 reliquiis beati Johannis et de miraculis quae ibi Dominus ostenderat, 
 legates sues Morknnam direxit, qui ecclesiam inibi fabricarent, cum 
 circumjacentibus episcopis et comitibus, ubi reliquias beati Johannis 
 Baptistse reponeret, eamque perfectam episcopo Viennensi, ad cujus 
 dioecesim pertinebat locus, sancto Isychio sacrare prsecepit. Synodum 
 vero postmodum in civitato Cabillonis congregare sanctorum episcoporum 
 fecit, et ibidem sanctum Felmasium episcopum Morigennce ab episcopo 
 Viennensi ordinatum primum constituit, et civitati Viennensi ipsam 
 Mauriennam ecclesiam cum consensu episcoporum subjectam fecit. 
 Ad quam ecclesiam Morigennensem, ubi beati Johannis Baptistse reliquias 
 posuerat, Secusiam civitatem jamdudum ab Italis acceptam cum omnibus 
 pagensibus ipsius loci subjectam fecit, et consensu etiam Romani Ponti- 
 ficis Viennensi ecclesise jure perenni episcopus civitatis et vici Mauri- 
 
 gmncB subditum esse decrevit Appendix to the Works of Gregory of 
 
 Tours. 
 
 3 The 'Morginno' of the Peutingerian Table should, perhaps, be 
 *Morginna' Vienna, in this Table, is written * Vigenna;' a further argu- 
 ment, if one were required, for the identify of 'Morienna' and 'Mori- 
 genna.' 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 
 I 
 
 177 
 
 Bernard were coincident. They went, by Lemincum (Le- 
 menc), across the Chartreuse mountains (probably by the 
 Mont du Chat), to Augustum (Aoste de St Genix); and 
 thence by way of Bergusium (Bourgoin), to Vienna Allohro- 
 gum, now Vienne. 
 
 It seems, then, from the agreement of two entirely 
 different lines of argument, that a Homan road must have 
 crossed the Little Mont Cenis. For that conclusion has 
 been drawn, in the first instance, from considerations quite 
 independent of the names of the stations ; the result hav- 
 ing^ been arrived at by applying to the Peutingerian table 
 an indispensable correction upon a system naturally sug- 
 gested by a previously discovered and corrected error in 
 the Table. The remarkable coincidences with respect to 
 the names of the Eoman stations seem to render it highly 
 improbable, that the conclusion previously drawn should 
 have been erroneous. 
 
 The grounds, on which it may be presumed that the 
 Mont Cenis was crossed by Julius Csesar, are derived 
 from the following passage in the first book de Bella 
 Gallico : 
 
 * Ipse (Caesar) in Italiam magnis itineribus contendit : 
 duasque ibi legiones conscribit ; et tres, quae circa Aqui- 
 leiam hiemabant, ex hibernis educit; et, qua proximum 
 iter in ulteriorem Galliam per Alpes erat, cum his quin- 
 que legionibus ire contendit. Ibi Centrones, et Garoceli, 
 et Caturiges, locis superioribus occupatis, itinere exercitum' 
 prohibere conantur. Compluribus his prseliis pulsis, ab 
 Ocelo, quod est citerioris provincise extremum, in fines 
 Vocontiorum ulterioris provincise die septimo pervenit- 
 inde in Allobrogum fines, ab Allobrogibus in Segusianos 
 exercitum ducit. Hi sunt extra provinciam trans Eho- 
 danum primi.* 
 
 Caesar's passage of the Alps was opposed by three 
 tribes; the Centrones, the Garoceli, and the Caturiges 
 The Centrones lived in the Tarentaise : the Caturiges bor- 
 dering on the Centrones, (who must be the Caturiges here 
 spoken of) in the Upper Maurienne. The third tribe, the 
 
 12 
 
178 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 Garoceli, is mentioned by Caesar only. Their name, does 
 not occur in Strabo, Pliny, or Ptolemy. The natural con- 
 clusion from these omissions is, that the Garoceli are 
 mentioned by those authors under some other name. They 
 appear, in fact, to be identical with the MeduUi, the city 
 of St Jean de Maurienne being mentioned in old docu- 
 ments as Sanctus Joannes Garocellius, a fact which seems 
 to fix the Garoceli in the Lower Maurienne, the country 
 of theMedulli^ ^ 
 
 Of the three tribes, then, who opposed Csesar's passage, 
 two lived in the Maurienne, and one in the Tarentaise. 
 Caesar must thus have crossed the Alps either by the Mont 
 Cenis or the Little St Bernard. 
 
 The Vocontian frontier, which Caesar would arrive at 
 by either of these passes, would be situated near the river 
 Drac, which seems to have formed the eastern boundary of 
 the Vocontii. He would, then, reach the banks of the 
 Drac, or the neighbourhood of Grenoble, on the seventh 
 day after leaving Ocelum. Now Ocelum lay, according to 
 Strabo, at a distance of 27 Roman miles below Scingo- 
 magus or Susa^; a measurement which would place it 
 
 ^ Incolas (Mauriennse) quamvis nonnulli putent eos esse, quos Pli- 
 nius ac Strabo Medullos appellaverunt, seu qui Brannovices C»sari 
 memorantur, nos tamen in ea sententia sumus, ut esse credamus ejus- 
 dem Csesaris Lib. i. Belli Gallici Garocellos, medios inter Vocontios ac 
 Centrones, id est, inter Montanos, Delphinates, et Tarentasios : eoque 
 magis huic sententise adhseremus, quo in antiquis tabulis ac monumentis, 
 S Joanms GarocelUi vocabulo appellatur legimus civitatis primarise, qua 
 de agimus, Ecclesiam Cathedralem.— Blaev. Theatrum Sahaudice. 
 
 The root of the name Garoceli is probably to be found in the Welsh 
 Gorui;eU 'Supreme, very high, lofty/ (Owen's Welsh Diet.) The Garo- 
 ceh were the inhabitants of some very lofty mountains. This circumstance 
 is perfectly in accordance with what Strabo says of the Medulli (Lib. iv 
 cap 6.)^ Mera bk OioKovriovs 2ck6pioc, kgI TpLKdptoi, Ka\ /xer* alroi,: Me- 
 dovXoi ocrrep r^, vylrrjXordra, ^^ovcrc Kopvcj^a,. Gor-u^el (an intensive of 
 U9eh high ) and lylrrjX^ordros have accurately the same meaning, and, 
 It IS possible, a common root. The lofty mountains of the Medulli 
 present themselves under an exceedingly imposing aspect when seen across 
 the Arc from the old road by Le Thyl and La Traverse 
 
 2 See Note B, at the end of the book, on the position of Ocelum, md 
 of hcmgomagus. 
 
 I 
 
 /' 
 
 /'/ 
 
 I/. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 179 
 
 3 Eoman miles below Avigliana, or near where Buttigliera 
 now stands. But for Caesar to perform, in seven days, the 
 journey from Buttigliera to Grenoble by the Little St 
 Bernard, would have been impossible: the distance is 
 evidently far too great. It must have been by the Mont 
 Cenis that he crossed the Alps ; for in this case the dis- 
 tance is found, as will be readily seen, to be about what 
 would have been previously expected. 
 
 From Ocelum to Susa the distance is 27 Eoman miles. 
 From Susa to St Jean de Maurienne, the distance is 
 58 Roman miles \ From St Jean de Maurienne to the road 
 leading up the Graisivaudan on the left bank of the Isere, 
 the distance by the mountain-road across the Col du 
 Glandon and the Col de la Coche may be roughly esti- 
 mated at 35 English, or 38 Roman miles. From thence to 
 Grenoble, the distance is about 12 English, or 13 Roman 
 miles. From Ocelum to Grenoble by this route the dis- 
 tance would thus be about 136 Roman miles 2. This 
 distance might have been performed by Cgesar in seven 
 days ; for he could have marched, when in haste, at the 
 rate of twenty Roman miles a day. That he was in haste, 
 is certain : and the mention of the seven days seems to 
 have been made with the view of proving the celerity 
 with which he marched. 
 
 When CaBsar says that he took the shortest road across 
 the Alps, he may probably mean, that instead of following, 
 from St Jean de Maurienne, the longer and easier route 
 by Aiguebelle, La Rochette, and Allevard, he took the 
 shorter way by the Col du Glandon and the Col de la 
 Coche, and was thus enabled to reach Grenoble in seven 
 days after leaving Ocelum. It is also not improbable, 
 that when on his way from Geneva into Italy to collect his 
 
 1 By the Peutingerian measurement, 5Q Roman miles. 
 
 2 Perhaps Caesar did not cross the Drac. Near the modern Grenoble, 
 he might perhaps be said to be * in finibus Vocontiorum.' Crossing the 
 Isere at Grenoble, he would arrive at the suburb of Grenoble lying on 
 the north of the river, and corresponding to the ancient Cularo, and thus 
 be * in finibus AUobrogum.' From thence, traversing the country of the 
 Allobroges to Vienne or Lyons, and crossing the Rhone, he would arrive 
 * in Segusianos.' 
 
180 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 forces, he had given some cause of offence to the Cen- 
 trones, whose country he would have skirted, or just 
 entered, if he followed, as is most likely, the road from 
 Geneva, by Annecy and Cesarches, to Aiguebelle and the 
 Mont Cenis. This may account for the Centrones joining 
 in the attack upon him on his return. At all events, they 
 had probably become aware that he intended to recross 
 the Alps into Ulterior Gaul, and were induced to join 
 their neighbours, the Caturiges and Garoceli, in an 
 attempt to resist his passaged 
 
 If the Mont Cenis was known to Julius Caesar, the road 
 over that mountain must necessarily have existed before 
 the time of Strabo. It may now therefore be concluded 
 with safety, that the pass of which Strabo speaks (Lib. iv. 
 cap. 6) between the country of the Medulli and that of the 
 Taurini*, was that of the Little Mont Cenis. (For this, and 
 not the Great Mont Cenis, is the line of road traced in the 
 Peutingerian table.) The distance, also, from Bramans to 
 Susa is, according to that Table, 25 Koman miles, or 
 
 1 The earliest route over the Mont Cenis would be that followed by 
 Hannibal, along the left bank of the Isere, and by the towns of Allevard, 
 Aiguebelle, and St Jean de Maurienne. Caesar, by taking the more diffi- 
 cult way from St Jean de Maurienne, across the Col du Glandon and the 
 Col de la Coche, would avoid the long circuit by the valley of La Rochette. 
 Up to this time, the Mont Cenis, the valley of the Arc, and the vale of 
 Graisivaudan, seem to have been used as the line of passage, both to the 
 neighbourhood of Valence, and also to Vienno and Lyons ; for it would 
 probably be to one of these last two cities that Caesar directed his 
 course, when he was on his way to the country of Segusiani. When the 
 route over the Chartreuse mountains was opened, the road of the Mont 
 Cenis was, as appears from the Peutingerian table, connected with that 
 road, and the way by Grenoble and the Graisivaudan no longer formed the 
 line of approach to Vienne. In the times of the empire, also, the road 
 from Gap to Die and Valence came into use, and, as would appear from 
 the Itineraries, displaced the Mont Conis from being the line of passage 
 between Valence and Turin. No route up the Graisivaudan is given in 
 the Itineraries, or traced in the Peutingerian table. None of the great 
 highways of the empire seem to have passed through it, though the 
 older roads in that valley may still have been more or less frequented. 
 
 ^ See above, p. 132, note 2. 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 181 
 
 200 stadia. The most accurate half-way point would 
 be where the road passed by the south-western corner of 
 the lake on the plateau of the Mont Cenis. From this 
 point, reckoning to Susa on the one side, and to Bramans 
 on the other, we should have the two lengths of one hun- 
 dred stadia each, which are mentioned by Strabo ^ 
 
 The concluding question connected with this enquiry 
 now arises. 
 
 Which of the two passes, the Little Mont Cenis and 
 the Mont Genevre, (a pass also known to Strabo) was 
 described by him, on the authority of Polybius, as the 
 pass through the country of the Taurini, which Hannibal 
 crossed^? For one only of these two passes was known 
 to Polybius, though both existed in the time of Strabo, 
 who seems to have introduced the words ^i/ 'AvvijSa^ StijXOel 
 for the purpose of definition, (and not as a mere casual 
 observation,) the pass which Hannibal crossed being neces- 
 sarily the most ancient of the two, and therefore the one 
 known to Polybius, of which Strabo wishes to speak. 
 
 Now the Alpine mountaineers were completely subdued 
 for the first time by Augustus. Cottius and the tribes 
 under his rule made their submission to the Romans, and 
 so escaped conquest ; and the Cottian territory seems to 
 have been increased, under Augustus, by the addition of 
 some of the conquered tribes, such as the Medulli, CatiZr 
 riges, and Vesubiani. Great alterations were effected by 
 Augustus in relation to the roads through the Alps. He 
 seems to have repaired the old roads and opened new 
 ones. It is in the reign of Augustus that the number 
 of the roads across the Alps would, in all probability, have 
 been increased. 
 
 The testimony of Caesar's Commentaries is particularly 
 valuable in indicating the roads which traversed the Alps 
 before the time of Augustus. For Caesar crossed the Alps 
 
 1 That Strabo mentions neither the Caturiges, nor the Segusini, in 
 connexion with this pass, seem omissions of no importance. 
 
 ^2 Terrapas 8" vjrep^daeis opofiaCft fiopoV dia Avyitou fxh t^v eyyiara tw 
 TvpprjpiK^ n€\dy€i ' €lra t^jp dia Tavplpa>p, r)p *Appl^as dujXOep' flra t^p 8ici 
 2aXao-(r(5»*' TfTaprrjp de, rfjp did 'Pair^p' dndcras Kprjppadeis.-^^traho, Lib. 
 IV. Cap. 6. 
 
182 
 
 ii;l 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 
 
 several times, and is himself his own historian, so that the 
 character of his evidence is perfect. Yet he only alludes 
 to two passes, the Mont Cenis, and the pass through the 
 country of the Veragri, (i.e.) the Great St Bernard ». He 
 does not appear ever to have crossed the Mont Genevre. 
 The Little Mont Cenis was thus, it is probable, the most 
 ancient pass of the two which led through the country of 
 the Taurini. 
 
 The route of the Mont Genevre, as given by Strabo, 
 led, by way of Embrun and the Durance, to Aries and 
 Nimes. The cross road from Gap, to Die and Valence, is 
 not mentioned by him. The route of the Mont Genevre, 
 as far as can be gathered from Strabo, was only known by 
 him as a road to Aries and Nimes, and as leading to the 
 Ehone by way of the Durance. The route of the Little 
 Mont Cenis passed through the country of the Medulli, 
 and would thus lead into the Graisivaudan, and reach the 
 Ehone by way of the Isere. 
 
 It was however, in the neighbourhood of the embou- 
 chure of the Isere, and not of the Durance, that Hannibal 
 struck off from the banks of the Ehone to cross the Alps. 
 It would therefore, on this account alone, be more pro- 
 bable, from what Strabo says, that Hannibal crossed the 
 Alps by the Little Mont Cenis, and not by the Mont 
 Genevre. 
 
 It appears then, on the whole, that the Little Mont 
 Cenis has the best claim to be considered as ' the pass 
 through the country of the Taurini, which Hannibal 
 crossed,' it being probably of more ancient date than the 
 Mont Genevre, and more consistent with what is known of 
 HannibaPs march along the Ehone. It is not contended, 
 that it is certain, from this passage in Strabo, and from 
 the fact of the antiquity of the road over the Little Mont 
 Cenis, that it was by this way that Hannibal crossed the 
 
 1 De Bello Gallico, Lib. in. These two passes appear to have been 
 known from the most remote historic times. The passes by which the 
 Gauls effected their earliest emigrations into Italy, according to Livy, 
 (v. 34, 35,) were two in number ; one througli the country of the Tau- 
 rini, and one by the Pennine Alps. 
 
 r 
 
 HANNIBAL'S PASSAGE OF THE ALPS. 183 
 
 Alps. The evidence, however, derived from the knowledge 
 of the Roman roads, is, as far as it goes, in favour of the 
 Little Mont Cenis ; and the amount of probability, what- 
 ever it be, which can be drawn from such testimony, rests 
 with that pass, and weakens a theory in support of any 
 other ; and thus the very grounds, on the supposed strength 
 of which, the claims of the Mont Cenis to be considered as 
 the pass which Hannibal crossed have been rejected, are 
 found, upon examination, to give additional force to those 
 claims. In fact, on whatever authority the enquiry into 
 Hannibal's route be based ; whether it be grounded upon 
 the narrative of Polybius or of Livy, or upon the know- 
 ledge of the ancient roads across the Alps ; the conclusion 
 reached is always the same, and the Little Mont Cenis is 
 pointed out, in every case, as the pass by which Hannibal 
 effected his descent into the plains of Italy. 
 
"^F" 
 
 APPENDIX. 
 
 ■li 
 
 Note A. 
 
 The following routes are taken from Wesseling's Itiwraric 
 Romanovum The first route is directly transcribed from the 
 text In the second route, three of the various readings given 
 m the notes have been adopted. The reading Cesuaria has been 
 preferred to Casuarta. The distance between Cesuaria and 
 Bautas is also, taken at xxx. instead of xviii. Roman miles. 
 Instead of Cenava, the reading Geneva has been adopted. The 
 modern places supposed to coincide with the ancient stations 
 have also been set down in the following tables: the reasons fo^ 
 such identification, m addition to those derived from the approxi- 
 mate coincidence of distances, will be afterwards given 
 
 ROUTE FROM IVREA TO VIENNE. 
 
 Ancient Stations. 
 
 Eporedia 
 Vitricium 
 Augusta Praetoria 
 Arebrigium 
 Bergintrum . . 
 
 Darantasia 
 
 Oblimum 
 
 Ad Publicanos 
 
 Mantala 
 
 Lemincum 
 
 Labiscone 
 
 Augustum 
 
 Bergusium 
 
 Vienna Vienne 
 
 Modem Places. 
 
 Ivrea 
 
 Verres 
 
 Aosta. 
 
 Arpi 
 
 Bourg St Maurice 
 
 Salins (near Moutiers) . . . . 
 
 Ablene 
 
 Le Fay 
 
 Maltaverne 
 
 Lemeiic (nearChambery). 
 doubtful 
 
 Aoste de St Genix 
 
 Bourgoin . . . , . 
 
 Distances 
 given in Ro- 
 man miles. 
 
 21 
 25 
 25 
 24 
 19 
 13 
 3 
 16 
 16 
 14 
 14 
 IQ 
 20 
 
 ROUTE FROM DARANTASIA TO GENEVA 
 
 Darantasia 
 
 Cesuaria , 
 
 Bautas 
 
 Geneva 
 
 Salins 
 
 Cesarches 
 
 Annecy-Ie-vieux 
 Geneva 
 
 24 
 30 
 25 
 
 13 
 
186 
 
 APPENDIX. 
 
 il 
 
 It will be convenient to consider the second of these routes 
 first. Darantasia must be either at Moutiers or at Salins In 
 addition to the argument drawn from the agreement of distances 
 the name of the station indicates that it was situated on the 
 Doron. the nver which joins the Isfere by Moutiers'. Besides, in 
 the Peutingerian table, Darantasia is placed 10 miles below the 
 station of Axima, clearly the modern Aime. M. Roche, of Mou- 
 tiers, Identifies Darantasia rather with Salins than with Moutiers. 
 The signification of the name of the Roman station makes Salins 
 also preferable. 
 
 The name of Cesarches, and its distance from Salins, point 
 
 t out as the Roman Cesuaria. Cesarches is a village lying on the 
 
 left bank of the Arly, a short distance above AlbertviUe The 
 
 inodern road from AlbertviUe, to Annecy and Geneva, lies on the 
 
 right bank of the Ally. 
 
 Bautas is generally placed at Annecy.le-vieux, which apnears to 
 be situated at about the right distance from C^sarches.Tnd also 
 from Geneva. The name of Bautas is perhaps identical with that 
 
 wesfof r' ^'" • " p^«f"^'''>) '^ P'^''*^ one or two miles to the 
 
 west of Carignano m Piedmont. Pautass in Piedmontese si-ni- 
 
 French ia Palud, which occurs in several places. The plain 
 where the modern Annecy is built may anciently have been a 
 marshy tract: or if the word Bautas be supposed endrelv iden- 
 tical with the Latin Palus, the lake of Annecy itse"f may be 
 pointed out under this designation 2, ^ 
 
 It has been already conjectured, that Arebrigium, on the first 
 NoHf °r •'^ P'''^^^ *t ^'•P' rather than at Prg St Didier. 
 Nothing else, with respect to this route, calls for notice before 
 arriving at Darantasia. ObUmum, (otherwise Obllonna), the first 
 
 at S«e "''""'"*'"' '*'*'"" ^''^''' ^°^^ ^^ "'^"'^ *"'' distance, 
 
 ,u ^^ /"•'"'''^nos % three miles below Oblimiim. Le Fay, on 
 the lef bank of the Isfere, is about that distance from AblSne 
 It would also be at about the right distance from Cesuaria or 
 Cesarches. The river was probably crossed at this place, and 
 
 fhe .^pTb^cani v"''" '"'''"'''■ '^"'^ ^^''^ ^^ *^- "' P-<- by 
 
 APPENDIX. 
 
 187 
 
 tiverloron tZJ . "" ^"'««-'-"' »«>"» P'-"ly 'o be ,he name of .he 
 nver />oro». The second pan probably remains in the name of the village of TKeu, 
 not far from AUevarJ. in Uauphin^. The Irish Teaffhais, • dwellinrplaees ! wHl be 
 
 ^°rDief!rb:rc;s tm:t:7: '" '"""-' '"""-'■ "«'• '^- '«^- 
 
 ' The M'elsh word • Fai- mean, ' an extreme ;' - Faig,' ' an extremity- or ■ furthest 
 
 The road, having crossed the Isbre at Le Fay, would pass, by 
 way of Ayton, and over the Arc, to Mantala or Maltaverne. 
 Recrossing the Is^re at Montm^lian, it would arrive at Lemincum 
 (Lemenc). The Chartreuse mountains were, in all probability, 
 traversed by the route of the Mont du Chat, from whence the road 
 descended to Yenne, and went by way of Aoste and Bourgoin 
 to Vienne. 
 
 Note B. 
 On the position of Ocelumy and of Scingomagus, 
 
 Ocelum is mentioned by Strabo (Lib. iv. Cap. i.) as being 
 situated at the extremity of the Cottian territory, at a distance 
 of 99 miles from Embrun, and 27 from Scingomagus, where Italy 
 began. 
 
 Ocelum is again mentioned by Strabo (Lib. v. Cap. i.) as 
 being reached from Placentia by a road along the Po and the 
 Dora, the Dora Susina being here plainly meant. This road 
 is said to be about 60 miles long. If reckoned from Placentia, 
 this must be a mistake for IGO : if reckoned from the point, 
 where the road would begin to follow up the Po, (i. e. from near 
 the modern Casale) the distance would be nearly correct, if 
 Ocelum lay not far from Avigliana. As the road from Placentia 
 would skirt the river Po first, Ocelum must have lain by the 
 Dora Susina. At Ocelum, Strabo says, the Alps and Gaul begin. 
 He had previously placed the beginning of Italy at Scingomagus. 
 It was consequently the Alps which began from Ocelum, and 
 then, after the mountains had been entered for some time, Gaul 
 began at Scingomagus. But perhaps Strabo here speaks of the 
 Cottian territory as part of Gaul, in which case it would be cor- 
 rect to say, that Gaul began from Ocelum. 
 
 Caesar speaks of Ocelum as the last place in the Roman 
 province of Cisalpine Gaul. 
 
 It appears then, that Ocelum stood in the province of Cis- 
 alpine Gaul, on the frontier of the Cottian land, at the foot of 
 the Alps, on the banks of the Dora Susina, and a distance of 
 99 miles from Embrun. 
 
 Ocelum is generally considered to be identical with Usseau 
 i.L the Val Pragelas. The similarity of the name is the only 
 reason for this supposition ^ Usseau is about 70 miles from 
 
 point.' Owen's Welsh Diet. The name of Le Fay might thus point to the fact of 
 its being the limit of the Provincia. 
 
 * Ocelum was a common name for a Celtic town. Besides the Usseau mentioned 
 in the text, the towns and villages of Ossola and Usseglio in Italy, Ussel and Issou- 
 
 / 
 
188 
 
 APPENDIX. 
 
 Embrun, by the road over the Mont Gen^vre and the Col de 
 Sestri^res, is not on the banks of the Dora Susina, is not at 
 the foot of the Alps, but far within them, and could hardly 
 have stood on the frontier of the Cottian land, but a long way 
 within its borders. 
 
 Avigliana (Ad Fines), according to the Itineraries, was 94 
 miles from Embrun, and lay on the banks of the Dora, at the foot 
 of the Alps, and on the frontier between the Cottian land and 
 Cisalpine Gaul. Ocelum must thus have been near Avigliana, if 
 not actually upon that spot. The modern Buttigliera perhaps 
 corresponds most nearly in position to the ancient Ocelum. 
 
 Scingomagus lay 27 miles from Ocelum, and therefore 72 from 
 Embrun. By the Itineraries, Susa was 70 miles from Embrun, 
 and, by the Peutingerian table, 72. According to Strabo, Italy 
 began at Scingomagus; according to the Jerusalem Itinerary, 
 Italy began at Susa. It seems thus quite evident, that Scingo- 
 magus, and Secusio or Susa, were the same place ^. 
 
 dun in France, and Ochiltree (i. e. Uchel-tre, * the high-town,' Chalmers' Caledonia) 
 in Scotland, all shew their connexion with the old name. The root of the word is 
 found in the Breton uc^hel and the Welsh ugel, both signifying high, towns being 
 commonly built upon heights, as defensible positions. The proper meaning of Oce- 
 lum seems to be a, fort or strong place. The Usseau of the Val Pragelas would pro- 
 bably have been one of those ocela : but the mere identity of name, when the appel- 
 lation was so common, is not sufficient to compensate the complete discrepancy, in 
 point of situation, between Usseau and the Ocelum of Cffisar and Strabo. 
 
 ' The termination magus is not uncommon. Caturiges (now Charges) is called 
 Catorigomagus in the Peutingerian table. In Rotomagus (Rouen), JVoviomagus 
 (Speyer), and Borbetomagus (Worms), it also appears. This termination seems to be 
 the Latin form of the Celtic magh^ ' a field,' a word apparently preserved in the names 
 of the Irish towns oiOmagh and Armagh. (See Diefenbach's Celticay i. 77, Art. 105.) 
 In the first part of the word Scingo-maguSf ( S/ctyyo^iayos) the name of the tribe 
 Segusini or Secusini is probably involved. 
 
t 
 
November 1853. 
 
 A LIST OF BOOKS 
 
 PUBLISHED BY J. DEIGHTON, 
 
 CAMBRIDGE, 
 
 AGENT TO THE UNIVERSITY. 
 
 CLASSICS. 
 
 JESCHYLUS—TRAGCEDIJE, Recensuit, emendavit, ex- 
 
 planavit, et brevibus Notis instruxit F. A. PALEY, A.M., olim Coll. 
 Div. Johan. Cant. 2 vols. gvo. \l. 4*. 
 
 Or Separately^ 
 
 s. d. 
 Oresteia. (Agamemnon, 
 
 Choephori, Eumenides) 7 6 
 Aframemnon (2nd Edit.)... 4 6 
 
 Chocjihori „.. 3 6 
 
 Eumenides (2nd Edit.) ... 4 
 
 _. s. d. 
 
 Persae 3 g 
 
 Prometheus Vinctus .......!.!!.... 4 
 
 Septem contra Thebas, et Frag.1 
 
 menta 5 g 
 
 Supplices (2nd Edit.) ..!..'!.'.'.'..'.*.'![ 4 6 
 
 DISSERTATIONS on the EUMENIDES of ^SCHYLVS: 
 
 with Critical Remarks and an Appendix. Translated from the German 
 of C. O. MiiLLEB. Second Edition. q^, 6d, 
 
 PASSAGES m PROSE and VERSE from ENGLISH 
 
 AUTHORS for Translation into Greek and Latin ; together with selected 
 Passages from Greek and Latin Authors for Translation into English : 
 forming a regular course of Exercises in Classical Composition. By 
 the Rev. H. ALFORD, M.A., Vicar of Wymeswold, late Fellow of 
 Trinity College, Cambridge. gvo. 6«. 
 
2 CLASSICS. 
 
 GEMS of LATIN POETRY. With Translations, selected 
 
 and illustrated by A. AMOS, Esq., Author of the Great Oyer of Poison- 
 ing, &c. &c. 8vo. 12*. 
 
 THE POLITICS OF ARISTOTLE; with Notes. By R. 
 
 CONGREVE, M.A., Fellow and Tutor of Wadham College, Oxford. 
 
 Preparing, 
 
 ARUN DINES OAML SIvc Mnsarum Cantabrlglcnsiura 
 
 Lusus Canori; coUegit atque edidit H. DRURY, A.M. Editio quarta. 
 
 8vo. 12a. 
 
 ARISTOTLE^ a Life of: including a Critical Discussion 
 
 of some Questions of Literary History connected with his Works. By 
 J. W. BLAKESLEY, M.A., late FeUow and Tutor of Trinity College, 
 Cambridge. 8vo. S*. 6rf. 
 
 DEMOSTHENES' SELECT PRIVATE ORATIONS, after 
 
 the Text of Dindorf, with the various Readings of Reiske and Bekker. 
 With English Notes, by the Rev. C. T. PENROSE, M.A. For the use 
 of Schools. Second Edition. • 4», 
 
 DEMOSTHENES DE FALSA LEGATIONE, By E. 
 
 SHILLETO, M.A., Trinity College, Cambridge, and Classical Lecturer 
 at King's College. Second Edition, carefully revised. Ready, 
 
 DEMOSTHENES, Translation of Select Speeches of, with 
 
 Notes. By C. R, KENNEDY, M.A., Trinity College, Cambridge. 
 
 12mo. 9*. 
 
 TOEPIAHS KATA AHMO20ENOTS. The Oration of Hy- 
 
 perides against Demosthenes, respecting the Treasure of Harpalus. The 
 Fragments of the Greek Text, now first Edited from the Facsimile of 
 the MS. discovered at Egyptian Thebes in 1847; together with other 
 Fragments of the same Oration cited in Ancient Writers. With a Pre- 
 liminary Dissertation and Notes, and a Facsimile of a portion of the MS. 
 By C. BABINGTON, M.A., Fellow of St. John's College, Cambridge. 
 
 4to. 6*. Qd. 
 
 CLASSICS. 3 
 
 'li^AISTmNOS EPXEIPIAION OEPI METFilN KAI nOIIIMATaN. 
 —The ENKHEIRIDION of HEHFAISTIOWN concerning METRES 
 and POEMS. Translated into English, and illustrated by Notes and 
 a Rythmical Notation; with Prolegomena on Rythm and Accent. By 
 T. F. BARHAM, M.B., formerly of Queens' College, Cambridge; Mem- 
 ber of the Royal College of Physicians, London. 8vo. 85. Qd. 
 
 VARRONIANUS, A Critical and Historical Introduction to 
 
 the Philological Study of the Latin Language. By the Rev. J. W. 
 DONALDSON, D.D., Head-Master of Bury School, and formerly 
 Fellow of Trinity College, Cambridge. Second Edition, considerably 
 enlarged. 8vo. Us. 
 
 EURIPIDIS ALCESTIS, Ad fidem Manuscriptorum ac 
 
 veterum editionum emendavit et Annotationibus instruxit J. H. MONK, 
 S.T.P., Grajcarum Literarum apud Cantabrigienses olim Professor Regius. 
 Editio Septima. 8vo. 4s. Qd. 
 
 EURIPIDIS HIPPOLYTUS CORONIFER. Ad fidem 
 
 Manuscriptorum ac veterum editionum emendavit et Annotationibus 
 instruxit J. H. MONK, S.T.P., Grajcarum Literarum apud Cantabri- 
 gienses olim Professor Regius. Editio Quinta, recensita et emendata. 
 
 8vo. 5s. 
 
 EURIPIDIS TRA G (EDIJS Priores Quatuor, ad fidem Ma- 
 nuscriptorum emendata) et brevibus Notis emendationum potissimum 
 rationes reddentibus mstructai. Edidit R. PORSON, A.M., &c., re- 
 censuit suasque notulas subjecit J. SCHOLEFIELD. Editio Tertia. 
 
 8vo. 105. Qd, 
 
 FOLIOR UM SIL VULA : a Selection of Passages for Trans- 
 lation into Greek and Latin Verse, mainly from the University and 
 College Examination Papers ; Edited by H. A. HOLDEN, M.A., Vice- 
 Principal of Cheltenham College, Fellow and late Assistant Tutor of 
 Trinity College, Cambridge. Post 8vo. 7*. 
 
 By the same. 
 
 FOLIORUM CENTURIjE. Selections for Translation into 
 
 Latin and Greek Prose, chiefly from the University and College Ex- 
 amination Papers. I*ost 8vo. T*. 
 
CLASSICS. 
 
 CLASSICS. 
 
 GREEK TRAGIC SENARII, Progressive Exercises in, 
 
 followed by a Selection from the Greek Verses of Shrewsbury School, 
 and prefaced by a short Account of the Iambic Metre and Style of 
 Greek Tragedy. By the Rev. B. H. KENNEDY, D.D., Prebendary 
 of Lichfield, and Head- Master of Shrewsbury School, For the use of 
 Schools and Private Students. Second Edition, altered and revised. 
 
 8vo. S«. 
 
 TITUS LIVIUS, with English Notes, Marginal Eeferenecs, 
 
 and various Readings. By C. W. STOCKER, D.D., late Fellow of 
 St. John's CoUege, Oxford. Vols. I. and II., in 4 Parts, 12«. each. 
 
 CORNELII TACITI OPERA, Ad Codices antiquissimos 
 
 exacta et emendata, Commentario critico et exegetico illustrata. Edidit 
 F. RITTER, Prof. Bonnensis. 4 vols. 8vo. \l, 8«. 
 
 A few copies printed on thick Vellum paper, imp. Svo. 4/. 4*. 
 
 The THEATRE of the GREEKS, A series of Papers re- 
 
 lating to the History and Criticism of the Greek Drama. With a new 
 Introduction and other alterations. By J. W. DONALDSON, D.D., 
 Head-Master of Bury St. Edmund's Grammar School. Sixth Edition. 
 
 Svo. 15*. 
 
 PHILOLOGICAL MUSEUM, 
 
 2 vols. 
 
 8vo. reduced to 12«. 
 
 BCHLEIERMACHER' S INTRODUCTIONS TO THE 
 
 DIALOGUES OF PLATO. Translated from the German by W. 
 DOBSON, A.M., Fellow of Trinity College, Cambridge. 8vo. \2s,Qd, 
 
 M. A. PLAUTI AULULARIA. Ad fidem Codicum qui iu 
 
 Bibliotheca Musei Britannici exstant aliorumque nonnullorum recensuit, 
 Notisque et Glossario locuplcte instruxit J. HILDYARD, A.M., Coll. 
 Christi apud Cantab. Socius. Editio altera. Svo. 7s. 6d. 
 
 M. A. PLA UTI MEN^CHMEL Ad fidem Codicum qui in 
 
 Bibliotheca Musei Britannici exstant aliorumque nonnullorum recensuit, 
 Notisque et Glossario locuplete instruxit, J. HILDYARD, A.M., etc. 
 Editio altera. ^vo. 7«. 6rf. 
 
 PROPERTIUS. With English Notes; a Preface on the State 
 
 of Latin Scholarship ; and copious Indices. By F. A. PALEY, Editor 
 of iEschylus. 8vo. 10*. 6d, 
 
 THEOCRITUS. Codicum Manuscriptorum ope recensuit et 
 
 emendavit C. WORDSWORTH, S.T.P., Scholae Harroviensis Magister, 
 nuper Coll. SS. Trin. Cant. Socius et Academia) Orator Publicus. 
 
 Svo. 135. 6d, 
 
 A few copies on large paper, 4to. II. 10s. 
 
 THUCYDIDES, The History of the Peloponnesian War: 
 
 illustrated by Maps taken entirely from actual Surveys. With Notes, 
 chiefly Historical and Geographical. By T. ARNOLD, D.D. Third 
 Edition. 3 vols. Svo. \l. 10*. 
 
 THUCYDIDES The History of the Peloponnesian War: 
 
 the Text of ARNOLD, with his Argument. The Indexes now first 
 adapted to his Sections, and the Greek Index greatly enlarged. By 
 the Rev. R. P. G. TIDDEMAN, M.A., of Magdalene College, Oxford. 
 
 Svo. 125. 
 
 SKETCH of the RISE and PROGRESS of ROMAN 
 
 LITERATURE. Chiefly from the Works of Bahr and Bernhardi. 
 By W. HAIG BROWN, M.A., Fellow and Tutor of Pembroke College, 
 Cambridge. Preparing, 
 
 SOPHOCLES, With Notes Critical and Explanatory, adapted 
 
 to the use of Schools and Universities. By T. MITCHELL, A.M., late 
 Fellow of Sidney Sussex College, Cambridge. 2 vols. Svo. II. Ss. 
 
 Or the Plays scparatelyy 5s, each. 
 
 PR^LECTIO IN PLATONEM, Habita in Scholis Publicis 
 
 Cantabrigiensibus vii kal Mai. a.d. MDCCCLIII. Auctore GULIELMO 
 HEPWORTH THOMPSON, M.A., Coll. SS. Trin. Socio, Gra^c. Litt. 
 Prof. Reg. Prqjariiiff. 
 
MISCELLANEOUS WORKS. 
 
 
 MISCELLANEOUS WOllKS. 
 
 PICTORIAL ILLUSTRATIONS of the CATALOGUE of 
 
 MANUSCRIPTS in Gonville and Caius CoUege Library. Selected by 
 the Editor of that Work. Being Fac-Similes of Illumination, Text, and 
 Autograph, done in Lithograph, 4to. size, to be accompanied with two or 
 three sheets of Letterpress Description in 8vo., as Companion to the 
 published Catalogue. Price £1. 4«. 
 
 A few copies may be had, of which the colouring of the Plates is more 
 highly jBnished; Price 1/. lO*. 
 
 TWO Introductory LECTURES upon ARCIIJEOLOGY^ 
 
 delivered in the University of Cambridge. By the Kev. J. H. MARSDEN, 
 B.D., Disney Professor of Archaeology. 8vo. clothy 2s. Qd. 
 
 LECTURES on the HISTORY of MORAL PHILOSOPHY 
 
 in England. By W. WHEWELL, D.D., Master of Trinity College, 
 Cambridge. 8yo. 8». 
 
 Th^ HISTORY of the JEWS in SPAIN, from the time 
 
 of their Settlement in that country till the Commencement of the 
 present Century. Written, and illustrated with divers extremely scarce 
 Documents, by DON ADOLPHO DE CASTRO. Cadiz, 1847. 
 
 Translated by the Rev. Edward D. G. M. Kirwan, M.A., Fellow of 
 King's College, Cambridge. Crown 8vo. 6s. 
 
 HISTORY of ROME. By T. Arnold, D.D., late Regius 
 
 Professor of Modern History in the University of Oxford. 3 vols. 8vo. 
 Vol. I., 16s.; Vol. II., 18*.; Vol. III., 14*. 
 
 ! 
 
 HISTORY of the LATE ROMAN COMMONWEALTH, 
 
 from the end of the Second Punic War, to the Death of Julius Caisar, 
 and of the Reign of Augustus ; with a Life of Trajan. By T. ARNOLD, 
 D.D., &c. 2 vols. 8vo. 1^. 8«. 
 
 CAMBRIDGE UNIVERSITY CALENDAR for 1853. 
 
 Continued Annually. 
 
 CAMBRIDGE UNIVERSITY ALMANAC for the Year 
 
 1853. Embellished with a Line Engraving, by Mr. E. CHALLIS, of a 
 View of Cambridge from the Castle-Hill, from a Drawing by B. RUDGE. 
 {Continued Annualli/.) 4,3, Qd. 
 
 GRADUATI CANTABRIGIENSES : sive Catalogus eorum 
 
 quos ab anno 1760 usque ad lO™ Octr. 1846, Gradu quocunque ornavit 
 Academia. Cura J. ROMILLY, A.M., Coll. Trin. Socii atque Academise 
 Registrarii. Svo. IO5. 
 
 THE BRITISH PALEOZOIC ROCKS AND FOSSILS. 
 
 By Prof. SEDGWICK, and Prof. M'COY. Royal 4to. 
 
 Part II. Description of the British Palaeozoic Fossils added by 
 Professor Sedgwick to the Woodwardian Collection of the University 
 of Cambridge ; with Figures of the new and imperfectly known Species. 
 By F. M'COY, Professor of Geology and Mineralogy, Queen's College, 
 Belfast. 
 
 \st Fasciculus^ (Radiata and Articulata). 16^. 
 
 2nd Fasciculus y (Lower and Middle Palaeozoic Mollusca). IO5. 
 
 Srd Fasciculus^ completing the Palaeontology. In the Press, 
 
 MAKAMAT; or Ehetorical Anecdotes of Hariri of Basra, 
 
 translated from the Arabic into English Prose and Verse, and illustrated 
 with Annotations. By THEODORE PRESTON, M.A., Fellow of 
 Trinity College, Cambridge. Royal Svo. 18«.; large paper, 1/. 45. 
 
 ARCHITECTURAL NOTES on GERMAN CHURCHES; 
 
 with Notes written during an Architectural Tour in Picardy and Nor- 
 mandy. By W. WHEWELL, D.D., Master of Trinity College, Cam- 
 bridge. Third Edition. To which is added. Translation of Notes on 
 Churches of the Rhine, by M. F. De LASSAULX, Architectural In- 
 spector to the King of Prussia. Plates. Svo. 12*. 
 
-hA r- 
 
 8 
 
 MISCELLANEOUS WORKS. 
 
 BEMARKS on the ARCHITECTURE of the MIDDLE 
 
 AGES, especially in Italy. By R. WILLIS, M.A., Jacksonian Pro- 
 fessor of the University of Cambridge. PlaieSf lakoe paper. 
 
 Royal 8vo. 1^ Is, 
 
 t 
 
 Quarto Series. 
 
 No. I. A Catalogue of the Original Library of St. Catharine's Hall, 
 M.cccc.Lxxv. By the Rev. Professor Corrie, B.D. 1*. 6d. — No. II. 
 Abbreviata Cronica. By the Rev. J. J. Smith, M.A. 2s. 6d. — No. III. 
 An Account of the Consecration of Archbishop Parker. By the Rev. 
 J. Goodwin, B.D. 3*. 6d. — Nos. IV. and V. An application of He- 
 raldry to the illustration of University and Collegiate Antiquities. By 
 H. A. "VVooDHAM, M.A. 5s.; 4s. 6c?. — Nos. VI. and VIII. A Catalogue 
 of Manuscripts and Scarce Books in St. John's College Library. . By 
 the Rev. M. Cowie, M.A. 4s. 6d. — No. VII. A Description of the 
 Sextry Bam at Ely, lately demolished. By the Rev. Professor Willis, 
 M.A. 3s. — No. IX. Architectural Nomenclature of the Middle Ages. 
 By the Rev. Professor Willis, M.A. {Out of print.) — No. X. Roman 
 and Roman-British Remains at and near Shefford, Bedfordshire. By 
 Sir Henry Dryden, Bart., M.A. 6s. 6c?. — No. XI. Specimens of College 
 Plate. By the Rev. J. J. Smith, M.A. 15s.— No. XII. Roman-British 
 Remains. On the Materials of two Sepulchral Urns. By the Rev. 
 Professor Henslow, M.A. 4s. — No. XIII. Evangelia Augustini Gre- 
 goriana. By the Rev. J. Goodwin, B.D. £1. — No. XIV. Miscellaneous 
 Communications. By Messrs. A. W. Franks, C. W. Goodwin, and 
 J. O. Halliwell. 15s. — No. XV. An Historical Inquiry touching St. 
 Catherine of Alexandria, illustrated by a semi- Saxon Legend. By the 
 Rev. C. Hardwick, M.A. 12s. 
 
 Octavo Series. 
 
 No. I. Anglo-Saxon Legends of St. Andrew and St. Veronica. By C. W. 
 Goodwin, M.A. 2s. 6d. — No. II. Greco-Egyptian Fragment on Magic. 
 
 By C. W. Goodwin, M.A. 3s. Qd No. III. Ancient Cambridgeshire. 
 
 By C. C. Babington, M.A., F.R.S. 3*. 6rf. 
 
 Report ; and Communications, No. I. 1«. — Ditto, No. II. Is. 
 Index to Baker's Manuscripts. 7s. 6d, 
 
 \^ 
 
REDUCTION 
 
 RATIO 
 
 WITHIN 
 
EawT'aved >ty JJt C Walker J 
 
Supposed route of Baruuiyal Colnurvd Blue 
 
 High roadjrom Chanibery to Turin .^ .... Rf3L 
 
 High, r^ad Jrt>m Cwrenoble into Stwoj Purple 
 
 Carriupe roadjrom Grenoble to MimtmeUan hy 
 Oie l^ bcadc of the Ltere. 
 
 Carruufe road Jronx Goncelin to Allevard ._ £r<mrv 
 
 Old roads from the Graiaivaudan to I 
 
 \ 
 
 EnojrayeA "by 
 
 J* C Walker^ 
 
^ UNIV'^Il<-'' 
 
 BRARIES 
 
 m^ 
 
 - ,f 
 
 •-«.. 
 
 ^ 
 
 . 
 
 r 
 
 
 --;' 
 
 1 . 
 
 li 
 
 
 ' 
 
 Ifl 
 
 
 
 
 
 
 
 1 
 
 
 
 
 '''iB 
 
 
 
 
 .3^ 
 
 
 
 ^"U;; 
 
 ■-■-"* 
 
 
 
 ^t'!' 
 
 ' ■' . . ::: 
 
 
 
 ' — . 
 
 ■■at 
 
 
 ^ 
 
 ■;^^ 
 
 
 '6': 
 
 "t^^ 
 
 
 iV 
 
 is i ffi 
 
 
 J---. 
 
 «^^ 
 
 
 S' 
 
 ? 
 
 
 r : 
 
 I; 
 
 
 "" 
 
 ; 
 
 
 V, 
 
 •! SiM'F' ' "■ 
 
 
 ^ i 
 
 
 
 r? 
 
 
 
 : -5^ 
 
 
 
 ■ ,•" ■ 1 
 
 
 
 ■ '^ 
 
 
 
 
 ■ ^ ^tt 
 
 
 
 ■*• 
 
 I 
 
 
 ■ 
 
 Hf. 
 
 'i 
 
 
 
 
 
 ■ 
 
 Be 
 
 J 
 
 
 
 ^f iij 
 
 
 
 1 
 
 
 
 ^ 
 
 IV- 
 
 
 i2 
 
 
 
 .*! 
 
 r 
 
 
 •:' 
 
 
 
 ... 
 
 W£3HiiasKr, 
 
 
 « 
 
 |^^t3}j^«'*'^v-\ , 
 
 
 
 
 
 •-• 
 
 
 
 ?? 
 
 i 
 
 
 
 ' iii 
 
 
 :.: 
 
 1 
 
 
 "-^ 
 
 Ht:K 
 
 
 
 vM^isiii^''^ 
 
 
 rat 
 
 t 
 
 
 '.-1 
 
 W 
 
 
 
 1 
 
 
 .-J 
 
 ■a 
 
 s 
 
 • 
 
 H 
 
 |B_~uitri»ita 
 
 
 i;^ 
 
 a'"''' 
 
 
 -:■ 
 
 & 
 
 
 1 
 
 
 
 
 EL*3 
 
 ».*' 
 
 .t{ 
 
 t: 
 
 i 
 
 3DN 12 1957 
 

 
 ^#i 
 
 « )!« S-f^ 
 
 
 KS-"'- 
 
 )#• <f 
 
 .t '■ 
 
 t / 
 
 «, t Iff,* »■ 
 
 u ii; t^" 
 
 
 'i*.- 
 
 '*»M 
 
 h^ 
 
 ft 
 
 >• «s 
 
 
 
 > ■, • 
 
 
 U'.". 
 
 "'j:v^ 
 
 
 ■«'4 
 
 :';f5j;'J 
 
 
 
 '■ i ^ ■