COLUMBIA LIBRARIES OFFSITE HI Allll v.il Nil S STANDAW) HX641 34431 QM575 .K95 On neurokeratin ON NEUROKERATIN 575 \W \Y. KIHNE 1'ltOFrseon OK FHTBIOLOaY IS llh'DK.I.i'.KKii INlVKIt-MTY and R. II. (IIITTENDEN PBORUOB OK PHYSIOLOGICAL CfiXJdBTST IN YALE T'NIVERSITT REPRINTED FROM THE SEW YOBE MEDICAL JOURXAI. FUR FEliinWRY ?:? AND MARCH 1, 1890 COLUMBIA UNIV'FRSITY DEPARTMENT OF PHYSiOLOrY College of Physicians and Surgeons 437 west fifty ninth street NEW YORK AP NEW Y B K 1). A PPL ETON AXD COMPANY 1890 Q2LL2L Air Columbia (Bntoettfitp intijeCiiptfltaitork College of -pf)£Sictaits ano gmrgeons Htbrarp Digitized by the Internet Archive in 2010 with funding from Columbia University Libraries http://www.archive.org/details/onneurokeratinOOkh ON NEUROKERATIN By \Y. KUHNE PROFESSOR OF PHYSIOLOGY IN HEIDELBERG UNIVERSITY and R. II. CHITTENDEN PROFESSOR OF PUTSIOLOGICAL CHEMISTRY IN YALE UNIVERSITY EKPRIN1ED FROM THE SEW YOJBR MEDICAL ,/OCRXAL FOR FEBRUARY 22 AND MARCH 1, 1890 N E W YORK I). APPLET ON AND COMPANY 1890 Q A! 5 75" ON NEUROKERATIN.* Tiik name neurokeratin f was applied to the substance contained in mednllated nerves and in the central nerve organs which is insoluble in alcohol and ether, in gastric and pancreatic juice, and in dilute caustic potash. It is therefore a substance characterized by extreme insolubil- ity. <>nlv the tissues and organs of the nervous system and the horny structures of the epidermis yield such an in- soluble bodv, so resistant to the action of all solvents, ex- cept such as are powerful decomposing agents. The preparation of neurokeratin depends upon the re- moval from the nerve tissue of the peculiar histogenetic sub- stances present there, such as albumin, collagen, elastin, and nuclein, likewise upon the separation of all fat and of the so-called myeline substances — viz., protagon, lecithin, cere- brin, cholesterin, and possibly soaps, and also of the ex- tractive matters soluble in water common to all tissues. The finding of neurokeratin led to the discovery that the nerve medulla was more complex in its composition than had before been supposed, especially as the presence of pro- teid matter was proved at the same time. Furthermore, it * PuUlifshcd also in the Zeitsclirift fur Biologie, Band xxvi. f A. Ewald and \V. Kiiline, l*el>er einen neuen Bestandtheil des Xcrvoiisystcins, Yerhandl. d. naturhist. nicd. Yereins zu Heidelberg, n. V., Band i, 1877, p. 857. 4 ON NEUROKERATIN. was likewise ascertained that, in addition to the neurokera- tin of the medullary sheath, other sheaths of similar ma- terial were also present in the medulla. In spite of these facts, however, we shall in the present article, for the sake of retaining simple terms, follow the old use of the word myeline and apply to the bodies extracted by alcohol, ether, chloroform, benzol, or carbon disulphide the name of med- ullary or myeline substances, although the term is now used in a much narrower sense. In order to obtain neurokeratin it is necessary to re- move the medullary substances or myelines from the brain or nerve tissue, as well as to free the latter from their pe- culiar tissue framework by digestion and from nuclein by extraction with alkali. As has been previously stated in another place, it makes no difference in the result as to the order of treatment, a fact of considerable importance, since the preparation of neurokeratin, by first digesting the nerve tissue and then extracting the myeline substances, furnishes proof that the long extraction of the medullary bodies with alcohol, etc., does not generate the insoluble substance, a fact which affords evidence of the pre-existence of neuro- keratin. For the microscopic study of the form and ai'range- ment of neurokeratin the method of treatment commencing with the extraction of the myeline bodies will presumably remain the most common, although even for this purpose the reversed method of treatment can be used with equally good results ; for the preparation of the substance in large quantities, however, only the first method has hitherto been used. We have for our present purposes employed both methods, and in the course of our experiments have found that digestion of the tissue prior to extraction of the mye- line substances, even where large masses of brain are taken, is not only practicable, but possesses certain advantages. ON NEUROKERATIN. Neurokeratin from Brain. The material best adapted for use i< the human brain, mainly on account of its large size. It was always ob- tained from an early dissected cadaver well preserved in the cold, and was always Fresh and without trace of putre- factive odor. In every preparation the first step consisted in the separation of the pia mater under water, so as to remove as far as possible all traces of Mood. 1. Preparation by Digestion after Extraction of the Medullary Substance. — Two brains were beaten up with 4,800 c. c. of absolute alcohol, the mass pressed through a sieve, and, after standing twenty-four hours, collected in a cloth bag or filter and the fluid pressed out. The residue was rubbed to a fine paste with a like volume of fresh alco- hol, and, after forty-eight hours, again pressed out and washed with cold ether. The mass was then thorounhlv washed with hot ether in an extractor with inverted con- denser for twenty-four hours, after which it was dried in the air, then boiled with two litres of absolute alcohol, pressed through muslin and placed in a special extraction apparatus, in which it was kept in contact with boiling alcohol for forty-eight hours. This apparatus was so ar- ranged that with the help of a sieve the mass of tissue was at Hist sunk deep into the alcohol, then raised nearer the surface, and finally suspended above the boiling alcohol, so that at the last it was extracted solely by the warm fluid Bowing back from the inverted condenser. After thorough pressure, the mass, freed in this manner from the medullary substances, was boiled with two litres of water, in order to completely remove the adhering alcohol and at the same time to render the connective tissue more digestible. It was then placed in five litres of an extremely active gastric juice containing - 4 per cent, hydrochloric acid and warmed <5 ON NEUROKERATIN. at 40° C. for four days, after which it was thrown on a large filter connected with a pump and thoroughly washed with water after sucking out all of the filtrate. The slight acidity of the remaining tissue was neutralized with dilute alkali, after which it was subjected to a trypsin digestion, in which one litre of an active pancreatic juice was employed, prepared in the usual manner from one hundred grammes of dry pancreas and containing 0*5 per cent, of sodium carbonate. After five days' digestion at 40° C. the undi- gested residue was washed with water, placed in two litres of five-per-cent. sodinm carbonate solution, and boiled for half an hour. The residue, after cooling, was filtered off, washed with water, and placed in three litres of 0-5-per-cent. caustic potash, where it was allowed to remain for forty-eight hours at medium temperature, with frequent stirring, in order to dissolve the nuclein present. After decantation, filtration, and washing with water, the mass was acidified with 0*4- per-cent. hydrochloric acid and subjected to a pepsin diges- tion for several days in two litres of an active gastric juice. Tn this last digestion there was no noticeable shrinkage of the mass, but, after washing out the acid with water, we found that hot alcohol still dissolved considerable matter, accompanied by a swelling or softening of the substance, in consequence of which the solution filtered very slowly. This admixture, consisting presumably of cerebrin, was re- moved to better advantage by repeated boiling with alco- hol, to which a little glacial acetic acid had been added. Thereupon the mass was treated again with caustic potash, and this time for two days with a solution of five per cent, strength, after which it was washed with water, then with acetic acid, again with water, and finally with alcohol and ether. The preparation obtained in this manner, on being tested with pure pepsin and trypsin, was found to be entirely free ON NEUROKERATIN. 7 from substances capable of furnishing peptone, also free from digestible albumin?, likewise free from soluble L r(, l ; i- tin, since tannic acid added to the solution obtained l>\ boiling with water gave do turbidity, and also free from nuolein, since five- per cent, caustic potash solution dissolved nothing precipitable by acetic acid. We were not able, however, to completely remove all of the myelitic bodies, for apparently a trace of cerebrin still remained. Except in the preparation of small quantities, as in the quantitative determinations to be described later, we have not been able to obtain a sample of neurokeratin which would not give up some matter on long-continued boiling with alcohol, benzol, or chloroform, the fluid on evaporation invariably leaving a slight film or coat on the glass with a more or less fatty feeling. In view of the possibility of our having to deal with lime or magnesia soaps we have not neglected, as before mentioned, to boil with alcohol containing either glacial acetic acid or hydrochloric acid without, however, wholly reaching the desired end. All that can be accom- plished is to have the neurokeratin finally brittle in hot alcohol, so that the latter will filter easily and the prepara- tion itself not feel soft or fatty. In this we were occasion- ally more successful with the neurokeratin from ox brain than from human brain, hence possibly in the future ox brain will be the preferable material to work with. 2. Preparation by Digestion prior to Extraction of the Medullary Substances. — For this purpose it is advantageous not to employ more than one brain (about 1,300 grammes) in a single preparation. The tissue was at once rubbed up fine, with gradual addition of 0'4-per-cent. hydrochloric acid, and to this was finally added five times its volume of active gastric juice. The juice was always prepared by the self-digestion of one part of the mucous membrane of a pig's stomach, with twenty parts of 0-4-percent, hydrochloric g ON NEUROKERATIN. acid, and preserved in the cold after being moderately thy- raolized. As a rule, the brain-matter, after four days' di- gestion at 40° C, sank so completely to the bottom of the vessel that the excess of the juice could be decanted off, while the residue was collected on a filter, again rubbed up as fine as possible in a porcelain mortar, and once more warmed for four to six da\s with four litres of fresh gastric juice. In every case we have tested a sample of the gastric juice employed, in order to be sure of its activity, by try- ing its solvent power on blood-fibrin. Although the diges- tion periods were quite long, we never observed the forma- tion of any mold or decomposition through the growth of micro-organisms, though the clear filtrates frequently broke down on long standing after the thymol had nearly or com- pletely left them. After the second digestion, the brain-matter was again collected on a filter, and, after being washed, was either directly freed from mveline bodies or else subjected to a trypsin digestion (possibly unnecessary). In the latter case the mass was neutralized with sodium-carbonate solution, and then warmed for eight days at 40° C. with one litre of well-thymolized pancreatic juice (1 to 40) containing 0*5 per cent, of sodium carbonate. At the end of the digestion the whole mass was in the form of a soapy magma, and, as filtered portions on neutralization or by acidifying were not rendered turbid, the entire mixture was at once acidified with hydrochloric acid up to about 0*1 per cent., and then ^shaken with ether. The treatment with acid was necessary, as the ether would not separate well, if at all, from the alkaline fluid. The separation of lecithin, protagon, and cholestcrin by means of ether was accomplished by repeated shakings in a large separating funnel, and at the same time, also, the soluble digestion products were washed out, since the brain-matter under the ether formed such a compact on NEUROKERATIN. 9 layer. Boating on the acid fluid, thai the latter only Deeded to be drawn off and the operation repeated with fresh dilute acid and water. When tlii^ had been continued until a portion of the aqueous fluid was not rendered tnrbid by a large excess of alcohol, the pasty like mass under the ether was drawn off, mixed with alcohol until it could be easily filtered, boiled with 60-per-cent. alcohol for half an hour, and filtered on a hot-water funnel. The filtered fluid, at first clear, quickly became milky on cooling from the sepa- rated cerebrin. The brain residue was then repeatedly boiled with absolute alcohol, benzol, and chloroform, the filtrations being made with the aid of the hot-water funnel. In order to remove the nuclein from the much-shrunken preparation, it was treated for several days with about two litres of one-per-cent. sodium hydroxide solution, filtered, washed with the alkali until the washings were not rendered turbid by acid, then with dilute hydrochloric acid. Finally, the residue was boiled with alcohol containing a little hy- drochloric acid, then with the various solvents for the mye- line bodies, and at last thoroughly washed with ether. Whether it is possible by either one of the above meth- ods to prepare larger quantities of neurokeratin pure is un- certain. We need to remember, in considering the prepara- tion of this substance, that we are dealing with a body which can not be dissolved unaltered and then reprecipitated, but with one which can only be obtained as a residue, and, fur- ther, that, in spite of great patience, we have not been able, as before mentioned, to entirely remove the myeline bodies. A further difficulty is to be found in the visible admixture of insoluble pigments, for neurokeratin obtained from thegraa substance, or from the entire brain, has a distinct yellowish color, while that prepared from the white substance is color- less. Another slighl impurity is to be found in the presence of paper fibers, although we have avoided these as much as 10 ON NEUROKERATIN. possible, even at the expense of our product, by leaving considerable of the sticky substance adhering to the paper. This we further endeavored to reduce to a small micro- scopic residue by working the preparation, while still moist with alcohol and ether, on a glass plate with a blunt knife, in order to felt the fibers into little flocks, many of which tended to stick to the blade. Still more was removed, after drying and pulverizing the product, by sifting the powder through fine mull. We have also tried shaking and wash- ing the substance with heavy fluids, such as chloroform or carbon disulphide, but without result. The following analyses were made with products dried at 110° C. until of constant weight. Carbon and hydrogen were determined by combustion in oxygen gas in a tube filled with granular oxide of copper, with an anterior layer of chromate of lead and a roll of reduced copper. Nitro- gen was determined as gas * ; sulphur, according to Ham- marsten's method. I. Neurokeratin from two human brains, in which the medullary substances were first extracted, using only alcohol and ether, and the residue of tissue then digested : I. 0*5377 gramme of substance gave 0'3522 gramme H 2 = 7*27 per cent. H, and 1*0928 gramme C0 2 = 55*42 per cent. C. II. - 4389 gramme of substance gave 0*2850 gramme H 2 = 7*21 per cent. H, and 0*8926 gramme C0 2 = 55*46 per cent. C. III. 0*5250 gramme of substance gave 63*7 c. c. N at 19*6° C. and 754*9 mm. pressure = 14*13 per cent. N. IV. '52*78 gramme of substance gave 63*7 c. c. N at 19*6° C. and 761*4 mm. pressure = 14*18 per cent. N. V. 0*6852 gramme of substance, fused with KOH + KN0 3 , gave » 0*0930 gramme BaS0 4 = 1*86 per cent. S. VI. 0*5455 gramme of substance, fused with KOH + KN0 3 , gave 0*0744 gramme BaS0 4 — 1 *87 per cent. S. VII. 0*6597 gramme of substance gave 0*0075 gramme ash (calcium phosphate) = 1*13 per cent. * Zeitschrift fur Biologie, Bd. xix, p. 166. ON M'.l i;uKi;i;.\TIN. \{ VIII. 0*7247 grain n f sill >st ati<<- gave ihiil'.i:! gramme ash = L'28 per (rut. Ptrcentagt Composition <>/ tin- Ash-free Substance, \ range, (' 50 09 5613 56 -I I II 7'88 7-30 N 14:50 14-35 S 1ST 1 O . . 733 . . 14-32 88 1-H8 . . 20-36 ioo-oo II. As preparation No. 1 had a somewhat greasv feel- ing and yielded a little something to hot benzol, it was again thoroughly extracted with boiling benzol, chloroform, alcohol, and carbon disulphide, and an endeavor made to free it still more completely from paper fibers. After being dried at 110° C.,it was analyzed with the following results: I. 0*3817 gramme of substance gave 0-2468 gramme H a O = 7'18 per cent. H, and 0-7830 gramme C0 a = 5594 per cent. C. II. 03825 gramme of substance gave 0*2480 gramme H 3 () = 7*28 per cent. H, and 0-7806 gramme CO a = 5565 per cent. C. III. 03724 gramme of substance gave 0-2416 gramme H 2 — 7*20 per cent. H, and 0*7620 gramme C0 2 = 55 - 79 per cent C. IV. 0-3154 gramme of substance gave 35 - 9 c. c. N at 5 4° C. and 7543 mm. pressure = 1391 per cent. N. V. 0-398.") gramme of substance gave 45*4 c. c. N at 6-0° C. and 757'6 mm. pressure = 13-95 per cent. N. VI. 05 126 gramme of substance, fused with KOII + KX0 S , gave 00612 gramme BaS0« = 1-63 per cent. S. VII. 0-4987 gramme of substance, fused with KOII + K\0 3 , gave 00570 gramme BaS0« = 1'57 per cent. S. VIII. 0'5726 gramme of substance gave 0-0050 gramme ash = - 87 per cent. IX. 0-6166 gramme of substance gave - 0057 gramme ash = o-92 per cent. X. 0-4985 gramme of substance, fused with KOII 4- KN'O,, precipi- tated with molybdio solution, and finally with magnesia mixt- c 56-44 56-15 56-29 H '7-24 V-29 1-26 N S .... 12 ON NEUROKERATIN. ure, gave - 0041 gramme Mg 2 P 2 7 = - 22 per cent. P, or no more than the phosphorus of the calcium phosphate of which the ash was composed. Percentage Composition of the Ash-free Neurokeratin. Average. 56-29 .... 7-26 14-04 14-08 14-06 1-66 1-60 1-63 20-76 100-00 As is seen from the results, the more complete extraction of the cerebrin (?) makes no special difference in the per- centage of composition when contrasted with No. 1. III. Neurokeratin from the brain of a child eighteen months old, purified as in No. II; first digested and then the medullary substances extracted : I. 0-3854 gramme of substance gave - 2580 gramme H 2 = 7'43 per cent. H, and 0-7894 gramme C0 2 = 55'85 per cent. C. II. 0-3660 gramme of substance gave - 2447 gramme H 2 = 7'42 per cent. H, and - 7524 gramme C0 2 = 56'05 per cent. C. III. 03474 gramme of substance gave 36 - 2 c. c. N at 5'4° C. and 755'7 mm. pressure = 12 - 76 per cent. N. IV. 0-3414 gramme of substance gave 36 - 2 c. c. N at 6'4° C. and 755-0 mm. pressure = 12 - 92 per cent. N. V. 0-4923 gramme of substance, fused with KOH + KN0 3 , gave 0-0610 gramme BaS0 4 = 1'70 per cent. S. VI. 0-5807 gramme of substance, fused with KOH + KN0 3 , gave 0-0743 gramme BaS0 4 = 1 75 per cent. S. VII. 0-4897 gramme of substance gave - 0076 gramme ash = l - 57 per cent. VIII. 0-4784 gramme of substance gave - 0073 gramme ash = 1*53 per cent. IX. 0-4923 gramme of substance, fused with KOH + KN0 3 , precipi- tated with molybdic solution, and finally with magnesia mixt- ure, gave 0-0079 gramme Mg 2 P 2 7 = - 44 per cent. P, corre- sponding to the calcium phosphate of the ash. ON NEUROKERATIN. 13 I'rr<-rn/ii(/) CiiiiijMixitinn •>( tin Ash-l'm Su/ixtmirr. Average. C 5672 6«'92 56-82 II 755 7-54 7-54 N 12-98 18-12 18-04 S 17:; 1-78 175 <> 20-85 lUIIMHI IV. Neurokeratin from tlie brain of an adult; method of preparation same as in No. Ill : I. 0'4082 gramme of substance gave 02930 gramme H 3 = 7'97 per cent. II, and 0-8688 gramme C0 a = 58 04 per cent. 0. II. 157- gramme of substance gave 0*8276 gramme H 2 = 7'95 per cent. H, and 09725 gramme C0 2 = 58-00 per cent. C. III. 0*3149 gramme of substance gave 292 c. c. N at 4'6° C. and 7526 mm. pressure = 11 34 per cent. N. IV. 0-3856 gramme of substance gave 35 - 8 c. c. N at 5 - 6° C. and 759-9 mm. pressure = 11 42 per cent. N. V. 0*6109 gramme of substance, fused with KOH + KX0 3 , gave 0-0672 gramme BaS0< = 1*80 per cent. S. VI. 0*8663 gramme of substance, fused with KOH + KN0 3 , gave 0-0490 gramme BaS0 4 = 1'89 per cent. S. VII. 0*8662 gramme of substance, fused with KOH -f KX0 3 , gave 0*0482 gramme BaS0 4 = 1-86 per cent. S. VIII. 0*4853 gramme of substance gave 0'0036 gramme ash = 074 per cent. IX. 0-9788 gramme of substance gave 00073 gramme ash = 0*74 per cent. X. 0-5108 gramme of substance, fused with KOH + KX0 3 , precipi- tatfd with molvbdic solution, and finally with magnesia mixt- ure, gave 0*0022 gramme MggPgOi =0*12 per cent. P, corre- sponding to the phosphate of the ash. I 'it-rentage Composition of ili<- Ash-free Substance, Average. C 58-47 58-43 58-45 11 8-03 801 8*02 N 11*42 11*60 11-46 S L-88 1-91 1:89 1-87 20 20 100-00 14 ON NEUROKERATIN. V. Neurokeratin from the brain of a man twenty-one years old ; method of preparation same as in Nos. Ill and IV: I. - 5783 gramme of substance gave 0-3803 gramme H 2 = 7'30 per cent. H, and T1820 gramme C0 2 = 55 - 73 per cent. C. II. 0-3695 gramme of substance gave - 2463 gramme H 2 = 7'40 per cent. H, and - 7606 gramme C0 2 = 56-13 per ceDt. C. III. 0"3881 gramme of substance gave 0-2578 gramme H 2 = 7'38 per cent. H, and - 7965 gramme C0 2 = 55 - 96 per cent. C. IV. - 4720 gramme of substance gave 48 - l c. c. N at 5*0° C. and 755 - 8 mm. pressure — 12 - 50 per cent. N. V. 0*8833 gramme of substance gave 39 - 9 c. c. N at 6*4° C. and 756 - 3 mm. pressure = 12 - 71 per cent. N. VI. - 5160 gramme of substance, fused with KOH + KN0 3 , gave 0-0825 gramme BaS0 4 = 2~19 per cent. S. VII. - 4806 gramme of substance, fused with KOH + KN0 3 , gave 0-0768 gramme BaS0 4 = 2-19 per cent. S. VIII. 0-5401 gramme of substance gave - 0130 gramme ash = 2 - 40 per cent. IX. 0-2742 gramme of substance gave 0-0065 gramme ash = 2-37 per cent. X. 0-4805 gramme of substance, fused with KOH + KN0 3 , precipi- tated with molybdic solution, and finally with magnesia mixt- ure, gave 0-0056 gramme Mg 2 P 2 7 = - 32 per cent. P, corre- sponding to the calcium phosphate of the ash. Percentage Composition of the Ash-free Substance. c . . . 57-09 57-48 57-32 H 7-48 7-58 7-56 S .... 12-80 13-00 2-i Average. . 57-27 7-54 . 12-90 2-24 20-03 24 I. C 56-11 H 733 N 14-32 S 1-88 100-00 Neurokeratin. Hair- II. III. IV. v. keratin. 56-29 56-82 58-45 57-29 49-45 7-26 7-54 8-02 7-54 6-52 14-06 13 04 11-46 12-90 16-81 1-63 1-75 1-87 2-24 4-02 Ash 1-21 0-89 1-55 074 2-38 1-01 ON NEUROKERATIN. 15 Examination of the above given results shows many points of interest in spite of the individual variations, which were perhaps to be expected from the manner of prepara- tion. First of all we notice the absence of phosphorus, con- cerning which we have reported in detail, because it mighl naturally he expected that an indigestible body like neuro- keratin should contain some substance related to the nu- cleins, as, for example, a nuclein insoluble in caustic alka- lies. The abseuce of phosphorus, however, shows that such is not the case. Secondly, the low amount of sulphur is in- teresting, especially when compared with older data, which show 2'93 per cent, in the case of neurokeratin from the brains of oxen and in preparations which had been extracted with only 0*5-per-cent. caustic soda. In Nos. I and II the preparation was, to be sure, extracted with an alkali of ten times the above concentration, but of the three other specimens, all of which were treated with only one-per-cent. caustic alkali, Xos. Ill and IV do not show any notice- ably higher percentage. In preparation No. V, however, the amount of sulphur rises to 2'24 per cent., although it was prepared in exactly the same manner as the preceding preparations and with no less care ; still the higher amount of ash is strongly suggestive of some impurity. As to how much sulphur is to be attributed to the albumins and albu- minoids is obviously a difficult question, but no more diffi- cult to decide than in the case of keratin and neurokeratin, since both continuously give up sulphur to caustic alkali, and that without losing the lead reaction of the more loosely combined sulphur. Doubtless the percentage of residual sulphur in these bodies is more dependent upon the abso- lute amount of alkali employed, upon the temperature or upon the duration of the extraction, than upon the concen- tration of the fluid. As preparations I and II show a difference of 0'25 per IQ ON NEUROKERATIN. cent, of sulphur, we endeavored to ascertain whether the fluids used in the extraction of the myeline bodies contained any sulphur at the end of the process. On testing the resi- dues, however, with acetate of lead and potassium hydroxide, we failed to find any sulphur reaction. Preparation No. IT, which was more' thoroughly extracted with alcohol, ben- zol, etc., was, however, 0*32 per cent, poorer in ash than No. I. Of the other constituents of neurokeratin, carbon is some- what higher and nitrogen somewhat lower than the percent- ages contained in albuminous bodies. In this connection it seemed very desirable to compare the composition of ordinary keratin with that of neurokera- tin, and, as the horny substance of the epidermis had never been studied, when purified from albumin by the method employed by us for neurokeratin, we have subjected hair to this process, purifying it by digestion. With this end in view a quantity of white hair from rab- bits, weighing 500 grammes, was digested for a long time with gastric juice and afterward with pancreatic juice. The quantity did not apparently diminish any by this treatment, although the gastric juice became somewhat colored and the fluid gave a not inconsiderable flocculent turbidity on neu- tralization. After thorough washing with water, the moist and matted mass was next treated for twenty-four hours with seven litres of 0*5-per-cent. caustic-soda solution, after which it was collected on a cloth filter and thoroughly washed with cold and hot water. On testing a sample, it was found to still give up considerable sulphur to alkalies, consequently the extraction with alkali was repeated and in this case con- tinued for three days with five-per-cent. caustic-potash so- lution. The mixture was then greatly diluted with water, filtered through flannel, the residue washed with water, and lastly with 0*5 per cent, hydrochloric acid. The hair while ON NEUROKERATIN. 17 in the alkali appeared < j u it e yellow in color, but after the washing it became glistening white again, and, as il was now somewhat broken to pieces and evidently more brittle than before, we thought it advisable while in this condition to extract it once more very thoroughly with gastric juice. Much to our astonishment, however, we found, after warm- ing the mass with ten litres of gastric juice for twenty -four hours, that full four fifths of the substance had disappeared; on treating the residue, however, with fresh gastric juice, it was not visibly altered. Under the microscope the hair structure was still visible, although the pieces were very soft, somewhat swollen and pale, and fell apart readily into single flakes or scales. The mass was finally boiled with alcohol, washed with ether, and dried at 110° C. The following re- sults show its composition : White rabbits' hair, purified by digestion, by removal of the fat and by caustic soda : I. 0*5348 gramme of substance gave 0*3120 gramme H 2 = 6*48 per cent. H, and 0-9616 gramme CO Q = 4903 per cent. C. II. 03162 gramme of substance gave 0-1832 gramme 'H 2 = 6-43 per cent. II, and 0-6655 gramme C0 2 = 48 - 77 per cent. C. III. 0-5340 gramme of substance gave - 3115 gramme H 2 = 647 per cent. H, and 0*9605 gramme C0 2 = 4905 per cent. C. IV. 04603 gramme of substance gave 615 c. c. N at 56° C. and 763*8 mm. pressure = 1653 per cent. X. V. 0*5312 gramme of substance gave 72 - 6 c. c. N at 5*6° C. and 757*9 mm. pressure = 1677 per cent. X. VI. 05300 gramme of substance gave 71 "8 c. c. X at 56° C. and 7 "'7 9 mm. pressure = 16'62 per cent. X. VII. 0-5889 gramme of substance, fused with KOII + KNOj, gave 0*1712 gramme BaS0 4 = 3-99 per cent. S. VIII. 0-6805 gramme of substance, fused with KOII + KX0 3 , gave 0-1963 gramme BaSO« = 396 per cent. S. IX. 0-6919 gramme of substance gave 0*0072 gramme ash = 1*04 per cent. X. 0'5921 gramme of substance gave 0-oo.v.) gramme a>li = 099 per cent. ; average I 01 per cent, ash, with much silica. 18 ON NEUROKERATIN. Percentage Composition of the Ash-free Substance. Average. C 49-53 49-27 49-55 49 45 H. . . . 6-54 6-50 6-53 6'52 N 16-70 16-94 16-79 16-81 S 4-03 4-00 4-02 23-20 100-00 We do not care at the present time to mate any close comparison of these results with the analyses of keratin pre- viously recorded, and will only call attention to the fact that our results do not differ essentially from the older analyses of keratin from hair, except that our figures for carbon and nitrogen are somewhat lower; on the other hand, the dif- ferences between our keratin results and those for neuro- keratin are very great, especially in the case of sulphur, which in the hair amounts to four per cent, or more, and that too in spite of the energetic extraction with five-per-cent. caustic potash. In making a comparison, however, we need hardly consider the sulphur, which certainly, in part at least, comes from some admixed body, for the. differences between the carbon and nitrogen are such as to plainly show that hair keratin and neurokeratin can not be directly grouped to- gether chemically. Indeed, the above-stated analyses con- stitute a most urgent reason for considering anew the whole keratin question in order to clear up first of all the great dif- erences in composition between the different keratins from the epidermis, nails, horn, feathers, etc. — differences which have long been known to exist. The earlier statements regarding the comparative in- solubility of neurokeratin from ox brain we have been able to substantiate with neurokeratin from the human brain; for, while rabbits' hair placed in five-per-cent. caustic potash for four weeks entirely disappeared,, neurokeratin was hardly altered by this treatment, and, further, on boiling ON NEUROKERATIN. 19 with dilute sulphuric acid, it left :i much larger residue than did hair. In attempting to make a comparative determination of the absolute and relative amounts of tyrosine and leucine which could he obtained from neurokeratin and from hair we met with a somewhat surprising result. Neurokeratin, on being boiled with dilute sulphuric acid for thirty-six to forty-eight hours, \ ielded both bodies in abundance, the tyrosine predominating, as is also the case with ox horn when purified in the ordinary way or by digestion. With the purified rabbit's hair, on the other hand (purified by digestion), we were not able with five grammes, a quan- tity ordinarily more than sufficient, to obtain any trace of tyrosine whatever. There was an abundance of leucine, hut no tyrosine; even the fluid from which the tyrosine should have crystallized, if present, gave only a slight orange color on being tested by the Millon-IIoffmann test. White rabbit's hair, unpurified or simply purified by digestion, yielded tyrosine quite plainly — apparently less, however, than yielded by the same quantity of horn, and even less than from fibrin. On testing the material furnished by a hair dealer, from which our purified preparation was made, five grammes boiled with dilute acid failed to give hardly any better reaction for tyrosine than the purified product, and it was only by taking fifteen grammes of the material that we were able to obtain suffieiem tyrosine to recognize the crys- tals. It seems probable, therefore, in spite of the protesta- tions of the dealer to the contrary, that in the technical working of the hair some powerful agents must have been used, which for our purpose was certainly unfort: nate. In connection with our study of this insoluble and indi- gestible constituent of the nervous system, we have tried an experiment of some morphological interest with the cord of the lobster. We dissected out as clean as possible tin- 20 ON NEUKOKERATIN. cords, consisting of nerve fibers and ganglia, from several medium-sized lobsters, an operation of no great difficulty, and digested tbem directly with gastric juice and then with pancreatic juice, after which the more or less adherent resi- due was extracted with alcohol and ether, and finally with caustic soda. That the residue consisted of chitin and not of neurokeratin was shown by dissolving it in concentrated sulphuric acid, pouring the solution into boiling water, neutralizing, and testing its reducing action on alkaline copper solution. It gave a strong reduction, similar to grape sugar. Quantitative Determination of Neurokeratin. — The meth- od employed- for the separation and purification of this body can also be used in its quantitative determination. Aside from the long waiting, the process can be carried out quite simply and accurately, but it is naturally a somewhat tedious one. Only one filter needs to be used — viz., that upon which the substance is finally weighed — and, aside from the watch-glass upon which the original tissue is weighed, only one vessel is, as a rule, needed in the process. The method, however, needs to be modified somewhat in the case of the white substance of the brain, and also when ex- traction of the myeline bodies precedes the digestion. We have confined our determinations wholly to the nerves and brain of man ; both should be freed as much as possible, before weighing, from external connective tissue and vessels, and also be cut into small pieces with scissors. As in this process, and in the separation of the gray and white substance, it is impossible to avoid the loss of water by evaporation, and as the material should amount to about fifty grammes, the preliminary weight can seldom be exact. We have, however, weighed each quantity as rapidly and accurately as possible, ordinarily placing the greatest value on the final weighing of the prepared substance. In this OX NEUROKERATIN. 21 way the different determinations were made under exactly the same conditions, ami are hence Btrictly comparable. Nerves or portions of the gray brain substance were rinsed, with five to >i\ times their quantity of gastric juice, into a glass-stoppered separating funnel of Too to 900 c.c. capacity, and subjected in this vessel, with frequent vigorous shaking, to digestion at 40° C. for eight to fourteen days. By thi> treatment the mass became a^ finely divided as if it had been rubbed up in a mortar. In the case of the white substance of the corpus callosum, however, this treatment is not so successful, since the substance sticks in part so closely to the sides of the vessel above the fluid that it is more or less withdrawn from the digestive juice. In this case, there- fore, the material was rubbed quite tine with the gastric juice, and the digestion carried on in a beaker, and when thesub- Btance had settled to the bottom of the dish, which gener- ally occurred on the fourth day, the clear supernatant fluid H;h in great part poured off and the sediment again rubbed up with a new lot of fresh gastric juice. After the pepsin digestion was finished, the residue was, at the beginning of our work, subjected to the action of trypsin, a step which we later found to be unnecessary. For this purpose the thick, but more or less fluid, residue was neutralized with a moderately strong solution of sodium carbonate, and then treated with 50 c. c. of an artificial pan- creatic juice for eight days, at 40° C, after which it was made distinctly acid with hydrochloric acid and shaken with ether. As a rule, we were generally successful in drawing off the acid watery solution beneath the layer of undigested matter, cpiite clear, and usually also the wash fluids, which were composed of 0*4-per-cent. hydrochloric acid. If, however, the fluids did show a DOticeable tur- bidity, they weir passed through the weighed filter, while the main bulk of the undigested matter was -till kept in the 22 ON NEUROKERATIN. separating funnel. After being thoroughly washed, the residue of tissue, made into a magma with a little water, was united with the upper ethereal layer by the addition of al- cohol, and the now homogeneous fluid, together with the undissolved matter, was allowed to flow upon the filter which was placed in a hot-water funnel. . In order to avoid all loss, the separating funnel below the stop-cock must be cut quite short, so that the mouth of the funnel can be well rinsed by means of a wash bottle. The hot-water funnel, on account of danger from fire, had attached to it a long rubber tube, through which the alco- hol passed into a receptacle placed some distance away; Further, in order to facilitate the boiling of the fluid on the filter, the rubber tube was provided with a spring clip, which could be closed during the operation of boiling. After the substance had been thoroughly extracted with alcohol and was ready to be washed with ether, benzol, or carbon disulphide, the burner was naturally extinguished and the tube removed, likewise after extraction with chloro- form, which we have also frequently used. The various fluids were poured upon the substance boiling hot, while the warm-water funnel was still hot, and, from its large size, remained so for some time. With this arrangement it was comparatively easy to free the digestion residue completely from recognizable traces of myeline bodies — or, in other words, by this meth- od we were able to accomplish what we did not succeed in doing with larger amounts of neurokeratin. After cooling, the mass was treated for at least two days with one to two litres of one-per-cent. caustic soda for the removal of the nuclein. The alkali was allowed to flow over the residue on the filter, being poured back and forth until the nuclein was wholly extracted ; it was then washed with water, after- ward with dilute hydrochloric acid, again with water, both ON NEUROKERATIN. 23 cold and li<>t, and finally with alcohol and ether, after which it was dried on the tiller, first in t he air, t hen at 110° C, and weighed. These various determinations should convince those who do not now apparently believe in the coexistence within the nervous system of the myelitic bodies and of a substance resistanl to the action of the digestive fluids and of caustic alkalies that neurokeratin really does exist; for in the weighed residues from the nerve filters, in spite of all that they had passed through, we could plainly detect micro- scopically the hornv sheaths and the spongedike neuro- keratin in the peculiar knotty or gnarled structure of the fragments as well as in the separated short cylinders occa- sionally seen, in which might also be discovered a second still smaller cylinder of the inner horny sheath. The frag- ments were colored intensely blue by hematoxylin, while by Ileidenhain's coloring, with the supplementary action of chromates, they were stained a blackish blue. They likewise gave the lead reaction for sulphur. I. 25 grammes from both brachial plexites of a woman seventy-two years old gave 0"079 gramme of neurokeratin = CK316 per cent. II. 50 grammes of the cortex of the cerebellum of a man twenty-one years old pave 0-1560 gramme of neurokeratin = 0-312 per cent. III. 50 grammes of white substance (not quite free from gray) from the cerebrum of the same man gave T1217 gramme of neuro- keratin = 2243 per cent. IV. 50 grammes of the pure white substance from the corpus callosum of a man fifty-seven years old gave 1'451 gramme of neurokera- tin = 2902 per cent. V. 50 grammes of the pray substance (as free ns possible from white) from the cortical portion of the cerebrum of the same man gave 0"1635 gramme of neurokeratin = 0327 per cent. The gray substance was separated from the white by scraping instead of by cutting, a method which enabled us to avoid any recognizable quantity of white matter. 24 ON NEUROKERATIN. In this connection it is interesting to compare the re- sults of determinations in which a portion at least of the myeline bodies were removed prior to digestion. To this end the material (nerves and white matter from the cor- pus callosum) was first placed in cold alcohol, the nerves in small pieces, while the brain substance was rubbed up with the alcohol. This alcohol was gradually replaced by fresh quantities of absolute alcohol, with which the tissue was at last thoroughly extracted and then treated with ether. Much of the fluid could be poured off quite clear, so that only the residue remained for filtration, for which the weighed filter was at "once used. After the removal of the ether by means of alcohol, and the latter by cold and finally boiling water, the substance was rinsed from the fil- ter with gastric juice, and, after being rubbed quite fine, was digested with the fluid in a beaker for fourteen days. In order to avoid all loss, and at the same time to treat all the tissue alike, the filter in its funnel was filled with gas- tric juice, the end of the tube being closed, and this like- wise placed in the digestion oven. During the digestion the residue was several times ground in a mortar, precautions naturally being taken to prevent loss. When the digestion was finished, the material was again brought upon the filter and a second time treated as already desciibed for the complete extraction of the myeline bodies. Lastly, the nu- clein was removed by the usual method. Following are the results obtained : VI. 50 grammes of nerve from the brachial plexus and the sciatic nerve of man gave O3005 gramme of neurokeratin = - 601 per cent. VII. 50 grammes of white substance from the corpus callosum gave 1-286 grammes of neurokeratin = 2'572 per cent. We give these determinations alone, although many others were made more especially for the purpose of work- ON NEUROKERATIN. 25 in-- out the method. As is Been from the results, the sec- ond method gives for the nerves (VI) almost doable the amount of residue fonnd by the firsl method (I); tor the white matter, on the other hand (VII), only a little less residue was found than in IV, and but little more than in III, so that very likely in a larger number of determina- tions little difference would l>e found. There is no ques- tion in our minds as to the cause of the difference in the case of the nerves; it is in great part dependent upon the fact that a portion of the connective tissue becomes indi- gestible by the alcohol treatment, a vie* which we have proved by our microscopic study of the question. Another cause for the difference is to be found in the presence of blood in the nerves, which were noticeably red, so that the neurokeratin had more or less of a brownish color, and, in fact, contained quite a little iron. In the white substance of the corpus callosum, on the other hand, where the con- nective tissue is, without question, much reduced and where no blood is visible, these possible admixtures do not influ- ence the amount of indigestible residue. Further confirmation of this view is to be found in the determination of neurokeratin made by Josephine Cheva- lier in 1885, a piece of work which has apparently been wholly ignored by histologists. This investigator found in the fresh sciatic nerve of man 0*30 per cent.* of neuro- keratin — a result which shows close agreement with our finding of 0316 per cent. Bearing in mind, however, that this result of Chevalier's was obtained from nerve fibers which had been previously freed from myeline bodies, it would seem at first glance as if this method should have yielded double the amount found, in correspondence with our determination of O'OOl per cent. (Analysis Yl). The reason, however, is to be found in the fact that Chevalier, * Zeitschrift fur physiol, Clieni., x, p. 100. i 26 ON NEUROKERATIN. before digesting the myeline free tissue in gastric juice, re- moved the connective tissue which had become indigestible through the alcohol-ether treatment by heating it with water at 120° C. in a sealed tube for twelve hours. The distribution of neurokeratin in the brain is quite remarkable. In no portion of the mammalian brain is there gray substance to be found which does not contain medul- lated nerve fibers ; but in the gray cortex of the cerebrum the admixture is quite limited, and in the corpus collosum white nerve fibers are present, with extremely well devel- oped and thick, spongy structure, and but little true neurog- lia. The cortex of the cerebellum (compare Analysis II), on the other hand, is composed of gray matter in which many medullated nerve fibers are to be found. The small amount of neurokeratin, only 0*312 per cent. — somewhat smaller even than that found in the nerves — would well agree with this view. In the cortex of the cerebrum the content of neurokeratin is somewhat higher than in nerves — viz., about 0-011 per cent. — too high, probably, to be ex- plained on the ground of admixture of medullated nerve fibers. Thin sections of the cortex, freed from myeline substance, prepared and treated according to a method to be described later in connection with nerves, have shown us beyond question the presence of a fine, spongy network of horny matter, developed thickest directly under the pia. The richness of the white substance in neurokeratin is quite remarkable, the amount being about nine times that contained in the peripheral nerves. Hence the brain is naturally the best material from which, to prepare large quantities of this substance. The great shrinkage of the connective tissue and the presence of the sponge-like net- work of horny matter all through the medulla find their chemical expression in the great preponderance of neuro- keratin. ON NEUROKERATIN. 27 In order to comprehend fully the significance of the above percentages, they need to be considered in connec- tion with the amount of water and of the so-called myeline substances of the tissue. According to < 'he\ alter, the ner\ eta of man contain 3372 per cent, of dry Babstance, of which 10-55 parts are insoluble in alcohol and ether. Thia mye- line-free substance would therefore, according to Analysis I, contain 1-909 per cent, of neurokeratin. The amount of neurokeratin contained in the myeline-free nerve fibers, however, would be somewhat higher than tlii-, as the dry substance necessarily contains some accessory tissues. Che- valier places it at 3*07 per cent. For the human brain it is somewhat hazardous to estimate the content of water, since the well-known analyses of the gray cortex by v. Bibra, Birkner, Bourgoin, and Weissbach show differences of 82*25 to 84*97 per cent., and in the case of the white substance of 63*54 to 73*93 per cent, of water. The average of all the analyses would indicate a content for the gray substance of 84 per cent, in round numbers, and of 70 per cent, for the white substance. Determinations of the myeline bodies in both the gray and white matter have been made only by Petrowski,* and the material used in his experiments was ox brains. Converting his results, expressed in percentages of dry substance to the moist weight of the human brain, we find for the gray matter 10*15 per cent, of myeline-free substance, and for the white matter 8*59 per cent. On the basis of Analysis V, the former would therefore contain 3-22 per cent, of neurokeratin, while the white matter, ac- cording to Analysis TV, would contain the enormous quan- tity of 33*77 per cent, of neurokeratin. Of the cerebellum we have not been able to find any analysis whatever, but, assuming that its content of water and myeline bodies is the same as in the cortex of the cerebrum, then its mve- * Arch. f. d. ges. Physiol., vii, p. 367. 28 ON NEUROKERATIN. line-free cortex would contain (Analysis II) 3*07 per cent, of neurokeratin. Hence our results indicate : For rnyeline-free dry nerve substance.. . . 1*91 per cent, neurokeratin. For myeline-free dry gray substance ... . 3*22 " " For myeline-free dry white substance . . . 33 - 77 " " " The old estimate that neurokeratin was present in the whole brain to the extent of at least 15 to 20 per cent, of the myeline-free dry substance was, therefore, none too high. In order to assure ourselves that in these determinations there was no great error from the peculiarities of the meth- od, especially as at times the weight of the filter was quite large in proportion to the amount of the undigested residue (0*9 to 2-7 grammes), we have applied the method to sev- eral other organs, using for this purpose the kidneys and liver. Forty-five grammes of fresh liver, with the usual amount of blood, from a recently killed rabbit, was first digested with gastric and pancreatic juice, then treated with alcohol, ether, and one-per-cent. caustic soda. The final amount of residue obtained was 0-0263 gramme == 0*058 per cent., or a little less than one fifth the smallest amount of neurokeratin found in the cortex of the cerebellum. Ten grammes of dry substance from a calf's liver, almost free from blood, was rubbed up with alcohol, then extracted with ether, etc., and, after thorough digestion and washing with boiling water, left a residue so small as to be un- weighable. Thirteen grammes of fresh kidney from a rabbit, on being treated in the same manner, left a residue of 0*0092 gramme = 0*070 per cent. — somewhat larger than the residue left by the fresh liver, but less than one fourth the residue from ON NEUROKERATIN. 29 the cerebellum. The residue had a grayish look and be- came blackish brown on boiling with acetate of lead and caustic soda. As the kidneys were omitted in an earlier investigation * of the indigestible constituents of various organs, a further examination of this small residue would be desirable. Microscopic Detection of Neurokeratin in Nerve Fibers. — The following description is intended to serve princi- pally as a guide to the detection of this peculiar insoluble portion of the nerve fiber which remains after all else has been dissolved away, rather than as a statement of the natural form and arrangement of neurokeratin in the living fiber. Neurokeratin was first described as having the form of a double sheath, of which the outer one surrounds the medulla of the nerve under the sheath of Schwann, while the inner incloses the axis cylinder, both being joined to- gether by means of a peculiar knotty structure. It was not assumed that the knotty forms corresponded exactly with those pre-existent in life, but it was plainly stated f that the prolongations of the outer horny sheath extend to the iuner, under certain circumstances, in the shape of the well- known spindle aud funnel forms of the medullary sheath. When nerves are treated with alcohol, from the condition of things normally present in the medullary space, their ap- pearance counts for little, since the naturally thick axis cylinder of the fiber shrivels and the small medullary mantle or sheath becomes thicker, through the action of the alcohol. Further, bridge-like connections must extend between these two horny sheaths, and these might thereby be broken. Ob- viously, therefore, for the first step in the treatment of the * VerhandL d. oaturhist. med. Vereins zu Heidelberg, X. F., Band I, pp. 451-456. f Jbid., p. 461. 30 ON NEUROKERATIN. fiber, only those reagents can be employed which will retain the natural volume of the axis-cylinder — as, for example, osmic acid, which, however, we have not as yet used in con- junction with the treatment by digestion and extraction of the myeline bodies. Suggestions have indeed been made as to the actual way in which the keratin network is formed and the manner of its arrangement, but they relate to micro- scopical forms from which neither the myeline bodies nor the digestible matter of the nerve were completely extracted. We refer to the so-called neurokeratin funnel and spiral forms described by several Italian observers, and to the well- known appearance of transverse sections of the medulla in osmic-acid preparations of nerves. These, recently ob- served by Joseph, but not wholly comprehended in their complexity, were long known to us as beautiful, untinted figures, which make their appearance in transverse sections of mammalian nerves, within the stained (by osmic acid) mantle of the medulla. The preparations were most advantageously obtained by extraction of the myeline substance and digestion of the teased nerve on a glass slide. For this purpose it is of some importance what kind of animal the nerves are taken from. In the nerves of the frog, and in those from fishes likewise, certain differences are to be observed, so that the following description, except when otherwise specially stated, refers solely to mammalian nerves, and particularly to those from rabbits, which appear best adapted for the purpose. As in the preparation of neurokeratin in quantity, so in the microscopical preparation there are two methods of treatment, in one of which the extraction of the myeline substance precedes the digestion, while in the other the latter process is applied first. We think it best, however, to de- scribe somewhat in detail both processes, since histologists ON NEUBOKEB \n\ 31 appear to be divided into two groups concerning tliis ques- tion, the one maintaining that undigestiblc structnrea are to be found in nerve fibers only when the myeline bodies have been removed by alcohol prior to digestion, and thai when the order of treatment is reversed Bach structures are not to be seen; the other group, on the contrary, contending that it is impossible to find Buch andigestible Btrnctnres in nerve fibers under any circumstances whatever. We will firsl call attention to the methods of treatment common to both processes, and to some of the precautions necessary to lie taken. 1. The complete extraction of the myeline substance, the difficulties of which have already been dwelt upon, is not easily accomplished, even in microscopical preparations where longitudinal and transverse sections of the nerve are only 15 to 20 /*, thick. We have indeed found it so diffi- cult that we hnve heen impressed by the lack of any state- ment to this effect in the many writings on tliis and kindred subjects, in which this circumstance must be of consider- able importance. These residues of myeline bodies are found in the nerve fibers, either in the form of glistening, oval-shaped balls — which, from their size and distribution, might be taken for nuclei — or in the form of a symmetrical thickening of the keratin network, comparable to a layer of varnish, thus imparting to the structure more luster and rendering it more refractive. This residue of myeline mat- ter is in all cases composed of cerebrin, which is n<>t easily soluble even in boiling alcohol or benzol, but, on the con- trary, dissolves out slowly, quickly separating from the fluid as it cools. It is necessary, therefore, after the preliminary treatment of the preparation with cold alcohol and ether, to bod it with both alcohol and benzol for five to ten minutes, at least three times, and to pour off the fluid each time while it is still hot, for the cerebrin shows a great tendency 32 ON NEUROKERATIN. to separate and adhere to the particles of tissue, even when the fluid itself shows no sign of turbidity. 2. The digestion of the nerve fiber we have in some cases carried on in large glass dishes, while at other times we have used small test tubes 35 to 45 mm. long and 15 to 20 mm. wide, loosely corked and set in a metal framework in a warm chamber, the temperature of which ranged from 37° to 41° C. The tubes with their contents of 5 to 10 c. c. were shaken as often as practicable, taking care, however, that the sections or small pieces of nerve are not broken by too vigorous agitation. In fact, there should be a rotation of the contents of the tube rather than a shaking, thus pre- venting at the same time the preparations from coming in contact with the cork, to which they are inclined to adhere. As to the length of time necessary to continue the digestion, probably twenty four hours in nearly all cases would be suf- ficient; we have, however, generally continued the process much beyond this length of time, usually for a week, and in some cases have even extended the time to seven weeks. More important, however, than the element of time is the condition of the digestive fluid and the frequency with which it is changed. Five different kinds of gastric juice, representing as many different methods of preparation, were used on the same object, one after the other: 1, two gly- cerin mixtures, the first composed of -|-, the second of -| pepsin-glycerin (prepared by us and known to be of maxi- mum activity) ; 2, a hydrochloric-acid extract of the mucous membrane of a pig's stomach, prepared by self-digestion at 40° C ; 3, same as 2, only prepared in the cold, both with -X mucous membrane ; 4, same as 2, with fa mucous mem- brane. All of the mixtures contained 0*4 per cent. HO. Each of these digestive fluids, when warm, dissolved raw fibrin almost instantaneously even when they were a month or more old. According to our experience, artificial gastric juice ON NEUROKERATIN. 33 prepared like the above and preserved in loosely stoppered flasks in a cool place, with a trace of thymol added, remains good for an exceedingly long time without loss of digestive power. All of our experiments, however, have been du- plicated with non-thymolized gastric juice preserved in the cold air of winter, the digestions themselves being car- ried on in a warm chamber, the air of which was kept strongly permeated with thymol. The trypsin solution was the same in all experiments; it always contained 0'5-per- cent. sodium carbonate, was repeatedly filtered from the separated ty rosin, saturated with thymol at the temperature of the room, and digested raw fibrin very noticeably in five minutes, completely in ten minutes. It was not so power- ful in digestive action as it is possible to prepare such a fluid, but was far more active than was necessary for the purpose in view. At the end of each digestion a portion of the fluid was always tested with fibrin in order to be sure that its energy was unimpaired. When the digestion of the object is completed it be- comes in many cases a matter of some importance, owing to the small size of the preparation and its great tenderness, to transfer it properly to the glass slide for examination. This we have most successfully accomplished by placing the small test tube containing the object in a trough, filled with water, made of glass plates fused together, whereby the fine sedi- ment is made more plainly visible on the curved bottom of the glass, whence it can be fished out with a suitable pipette. The same method can be advantageously used when it is necessary to extract the preparation with caustic soda or other reagents, or to wash it with water. The ex- traction of the preparation with caustic alkali for the re- moval of nuclein we have invariably performed, although it does not alter the microscopical appearance of the object, since the nuclein in these cases comes entirely from the 34 ON NEUROKERATIN. residues of the nuclei. As a preserving medium we have employed very dilute glycerin, first staining the more ten- der objects with Delafield's hematoxylin. I. Nerves digested after Extraction of the Myeline Sub- stance. — The nerves are first stretched out by means of threads in small glass tubes, or else fastened upon corks and then hardened in cold alcohol. After twenty-four hours they are cut into small pieces from 1 to 3 cm. long, and then freed from myeline matter, as already explained, by boiling with alcohol and benzol and also with ether. From a mixture of alcohol and ether the pieces are transferred to a solution of celloidin of gradually increased strength. After the hardening of the celloidin in seventy-per-cent. alcohol the pieces are cut by means of the microtome into longi- tudinal and transverse sections 15 to 20 /x, thick, which are then further extracted with the above reagents. While in- closed in the celloidin lamellae the tissue can not be treated with digestive fluids to advantage, since the smallest quan- tity of this substance renders both gastric and pancreatic juice wholly inactive, which fact is to be attributed to the peculiarity possessed by nitrocellulose, discovered by Dani- lewsky,* of precipitating the digestive enzymes. We found that a few thin shavings of celloidin, washed simply with wa- ter and placed even in large quantities of the digestive fluid, freed it so completely from the enzyme that its digestive power was almost wholly impaired. In gastric juice it would appear at the first glance as if transverse sections of a nerve do really dissolve, for after a time it will be noticed that a fine hole appears in the celloidin lamella, where the nerve section is known to be. On close scrutiny, however, it will be found that the entire tissue still remains, but ren- dered more or less transparent on the edge by the swelling of the connective tissue in the acid. That the gastric juice * Virchow's Archiv, xxxv, p. 2*79. ON NEUROKERATIN. 35 has really lost all of its digestive power can he easiU >\\<>w n by placing in it a piece of fibrin, which will he seen to swell up and hecoinc more or less transparent without, however, dissolving. The nerve sections must therefore be entirely freed from celloidin, and this operation serves likewise to remove all remaining traces of myeline matter. For this purpose the sections are treated for twenty-four hours with alcohol and ether and then again extracted with hot alcohol, benzol, etc., as previously descrihed, after which they are freed from alcohol by washing with water and then subjected to digestion. The digestion can be carried out in four different ways : 1, with gastric juice ; 2, with pancreatic juice ; 3, with pan- creatic juice after the sections have been previously boiled with water ; 4, with pancreatic juice after previous treat- ment with gastric juice or dilute hydrochloric acid. The third and fourth methods give the same results as the first, since from the preliminary treatment the connective tissue is rendered wholly soluble in the trypsin solution. The sec- ond method, on the contrary, preserves the collagenous fibers entire, together with the neurokeratin. An exception, how- ever, is to be noticed in this connection in the nerves of the frog. As Aug. Ewald * has recently observed, the tendons of the frog, in distinction from those of the Mam ma! in, are decidedly altered by trypsin, and the same is true, though to a less extent, of the connective tissue contained in the nerves of the frog. After a long digestion with trypsin the nerve fibers are seen to fall almost entirely apart, and it ap- parently makes no difference whether the nerve is fresh or has been previously extracted with alcohol. In both cases the nerve, on treatment with warm trypsin solutions, softens and the fibers separate, while the collagenous fibers which bound them together, and which at first can be readily iso- * Zeitschrift fur Biologic, xxvi, p. 1. 36 ON NEUROKERATIN. lated, after a time disappear, and in their place is to be found some soft, slimy matter or a membranous mass, the latter coming from the thicker neurilemma. The usual differ- ence, therefore, in the character of the residue left by agas- tric and pancreatic digestion disappears in the case of the frog's nerves, a fact which has been of some advantage to us in our study of this matter, as will presently be seen. In the mammalia this order of things is in a certain sense reversed, for here the connective tissue is so altered by the agents employed for the extraction of the myeline substances that it is not completely broken up by gastric juice, even after previous boiling with water, or by trypsin digestion when treated as described in 3 and 4. Herein lies the probable cause of the different results obtained in the quantitative determination of neurokeratin, both in nerves and in brain tissue, when the extraction of the myeline bodies precedes the digestion process instead of following it. Furthermore, this explains why microscopical prep- arations of neurokeratin are more readily obtained and in better condition from nerves which have been previously extracted with alcohol, etc., prior to digestion, since under such circumstances there remains after the latter process something which holds together the residue of the fibers. At the same time such preparations of nerve fibers are seen to be free from any accompanying tissue of recogniza- ble structure ; that which lies between the fibers is hardly recognizable as a cementing substance, for it is only on the edges of the section that the almost homogeneous border of transparent matter can be detected, showing here much like an outer sheath, thick and transparent. Another con- stituent often noticed is broken residues of the red blood- corpuscles, which have been rendered indigestible by treat- ment with alcohol. They often occur in rows, corresponding to the digested. blood- vessels, in between the residues of the OS NEUROKERATIN. 37 nerve fibers. Of the nerve fibers themselves, many are com- pletely dissolved, aside fr<>m the residues already mentioned, and these especially show what the arrangement <>f the neurokeratin in the Hher is. They are extremely friable, and are reduced to cylindrical, wrinkled Btrings or cords, in which can frequently be seen an inner string, like a wick in a candle, while surrounding the latter is a rumpled mass uniting it with the outer covering. We have treated hun- dreds of Bnch sections, and in many cases have allowed them to lie for months in a large excess of thymolized <_ r, i— trie juice in small test tubes, and in no case have we tailed to find the neurokeratin residue, and this both in large and small nerves alike. Further, in every case we have proved the great strength of the gastric juice emploved by testing its action on fibrin which had been boiled in water and also treated with alcohol and ether. Preparations which have been treated with trypsin, after the manner given under 2, are less changed than the pre- ceding, and retain the natural form of the neurokeratin better, since in these all of the collagenous fibers, which hold the nerve fibers together, remain, and, furthermore, are not swollen as when treated with the acid gastric juice. In addition, the nerve fibers themselves are less deformed than when treated with acid fluids, since the swelling action of the acid on the connective tissue fibers tends to shorten longitudinal sections of nerves, and hence twists and crum- ples the nerve fibers themselves. Generally the nerve fibers lie in straight rows, showing for long stretches an outer shell, together with an inner string of the shrunken horny sheath of the axis cylinder and the connecting struct- ure. In many places, however, there may be found por- tions of fibers in which it is impossible to r < gnizc the individual parts of the horny sheaths, where the fiber re- sembles a chain of crumpled fragments, owing to the coin- 38 ON NEUROKERATIN. plete solution of the contents of the sheaths and the elastic contraction of the neurokeratin, for, when all of the sub- stance naturally filling up the space of the fiber is removed, no resistance is offered to the elasticity of the neurokeratin, and hence it falls together more or less as one mass. If mammalian nerves are boiled with water or treated with acid, previous to subjecting them to the action of trypsin after the method under 3 and 4, they show the same appearance as when digested in gastric juice. More- over, fibers which have been digested in gastric juice are not apparently further changed by after-treatment with pancreatic juice. There is, therefore, a certain residue or decomposition product of the mammalian connective tissue present in the nerves after extraction of the myeline sub- stances, which resists the action of both digestive fer- ments. II. Nerves digested prior to Extraction of the Myeline Substances. — The character of the residue remaining after this method of treatment has long been known, and from the very beginning much weight has been attached to its peculiarities. The process, however, like that employed in the preceding case, is not so easy but that considerable practice is necessary to avoid loss of the undigestible residue by its floating off from the glass slide, or to obtain good preparations from the brittle mass left at the comple- tion of the process. Moreover, the conditions are not as favorable here for obtaining the horny matter in its natural form as where the digestion follows the alcohol treatment, for in the latter process the fluid portion of the medulla, especially that containing the albuminous matter, is first coagulated and hardened. Then, when the myeline bodies are removed, there remains a horny mass covered over or penetrated by the coagulated albumin and showing a form identical with the compact structure long ago described by ox NEUROKERATIN. 39 Henle and Merkcl,* whicli many later observers have held as congruent with neurokeratin, a mistake which Ilumpf has already called attention to. However, this preliminary coagulation of the proteid matter has an advantage for the digestive process in that it helps to fix the neurokeratin, and, further, the coagulated albumin is only slowly softened by the digestive juice, so that for this reason it ie possible to obtain the microscopic forms in much better condition than in the other method. In the direct digestion of fresh nerves, on the other hand, only soft matter fills out and sur- rounds the neurokeratin, from which the latter can find little or no support, so that in the rapid digestion and solu- tion of the non-coagulated albumin the horny residues natu- rally fall together and appear extremely rumpled and twisted. On being digested with gastric juice, a very marked shortening and swelling occur in fresh nerves, the extent of which can be estimated from the corresponding overflow of the medulla from the mushroom-like end of each piece of nerve. When all of the albuminous matter has been rendered non-coagulable by digestion and the inyeline bodies extracted from the nerve, it must not be expected that the residue will show more than a confused appearance, for it is only by the greatest care in the hand- ling of the object under the cover-glass that it will be possible in some measure to arrange the parts for observa- tion. If fresh nerves are digested with gastric juice in small test tubes and the medullary substances then removed by the ordinary method, the residue will be decidedly broken and mixed, for when the cerebrin is dissolved out by boil- ing with alcohol and benzol, the pieces become by this treat- ment so brittle that on the boiling of the fluid they are * Zeitschr. f. rat. Med., xxxiv, p. 49, Table V, RgB. 20 and 21. 40 ON" NEUROKERATIN. broken apart — so much so indeed that they can not be taken hold of by forceps or needles, but can be caught only by decantation of the fluid and the use of wide pipettes. The residue thus obtained, however, always possesses the same characteristic properties, as any one can see who will take the necessary pains in manipulation — viz., insolubility in caustic soda, intense coloration with hematoxylin, non- digestibility in gastric and pancreatic juice, etc. A far better method of obtaining neurokeratin for mi- croscopic study, and one which yields the product in much better condition, consists in the following treatment : A piece of the sciatic nerve from a rabbit, 2 to 3 cm. long, and which has been digested for one to seven weeks in 100 c. c. of active gastric juice, is thoroughly washed with water, then placed in alcohol of gradually increased strength, then in ether, in alcohol-ether, and finally in celloidin of gradu- ally increased concentration. In making these transfers from one fluid to another it is not necessary to use a pipette, which might possibly crush the brittle residue of the nerve, but it is better accomplished by careful pouring of the fluids, and, when necessary, by cautious slipping of the bit of tissue. The piece of nerve is then hardened with the celloidin in 70-per-cent. alcohol and cut into sections with the microtome. Longitudinal sections of the nerve 3 to 5 mm. long are best adapted for study. The celloidin sec- tions are taken from the 70-per-cent. alcohol singly by means of pinchers and placed in absolute alcohol, where they are allowed to remain until they begin to soften on the edges. This may be considered as evidence that they are sufficiently dehydrated to be boiled in benzol, by means of which the remainder of the myeline bodies, especially the cerebrin, is completely dissolved without alteration of the celloidin. From the hot benzol the celloidin sections are placed in cold benzol, then again in absolute alcohol until BOKBRATDI 41 they show a taadei (ten, then in dilute : .naJ- ly in water and in gljeerin, in the latter of which the \ aration can be best examined. In this manner all and rnmhrag are avoided and a preparaJ - .tied which, while in great part composed of twisted and k; - ?, still s - many pieces which may be ea-ily r. nized as residues >>t" ner\e fibers, with the char joining of the outer and inner horny sheaths and ti rk of neurokeratin. In order te exclude all doul B "he eompleten — the extraction of the myeliue bodies, the celloidin. which still surrouuds the residue from the nerve, can be complete- ly removed. This is best accomplished, after the final wash- _ with alcohol, by dropping upon the softened section placed on a h. ' _ tss slide boiling alcohol, and then the residue with ether, care being taken to avoid its : ; ing away. Addition of 1 to-5-per-cent. car.- - will then show the character of the insoluble substauee. The best preparations are obviously obtained from regular 1. I _ tudinal sections of nerves, while trausverse and oblique - - lions are more apt to yield poorer and less satisfactory speci- mens. For mieroscopists we would on some accounts recom- mend the pancreatic instead of gastric digestion, in which the collagenous fibers take the place of celloidin in prevent- ing the eutire collapse of the nerve fiber in the after-treat- ment with alcohol, benzol, etc.. and in which, likewise, the volume and length of the fiber remain practically unaltered. Furthermore, mammalian nerves can remain in the warm trypsin solution as long aa desired, and that too without g ition of the mixture. The pieces can then be washed iu water, freed from myelitic bodies, spread upon appropri- I the fibers teased apart in the ordinary tray. One would hardly expect to rind, after this method of treat- 42 ON NEUROKERATIN. ment and in such an intermingled mass of connective-tissue fibers, any nerve fibers showing so plainly the horny sheaths and the spongy network of neurokeratin as many of these do. The final treatment of the preparation with caustic soda occasionally causes difficulty, but this can as a rule be carried on so easily under a cover-glass that there is little danger to be apprehended. There remains a single word to be added concerning the peculiar behavior of frog's nerves toward pancreatic juice. In the isolation of neurokeratin from such nerves by means of trypsin, doubtless many have obtained unsatis- factory results, since in this case the connective tissue dis- appears and the nerve fibers fall apart. On shaking or agi- tating such a preparation, after the myeline bodies have been removed, the greater part of the residue from the tis- sue will undoubtedly be broken and lost, although the splinters of neurokeratin still recognizable will not be de- prived of their characteristic appearance nor fail to show their insolubility in caustic alkali. This difference in be- havior toward trypsin is, however, of decided advantage, for it facilitates the penetration of the various solvents em- ployed, and further yields the nerve fibers isolated in a manner fully equal to the result obtained by gastric diges- tion, bat with the added advantage of their being far less deformed, since the slow decomposition of the connective tissue by trypsin is not accompanied by the swelling and shortening of the nerves produced by a pepsin-acid mixture. The thick medullated nerves from the head of certain fishes, such as the barbel, behave in a similar manner toward tryp- sin ; they likewise yield an insoluble neurokeratin similar to the above, but in more delicate forms and often very beautifully arranged. We were somewhat surprised, how- ever, on comparing the behavior of the epidermal cells from the skin of the head and from the fins of the barbel with ON NEUROKERATIN. 43 those from mammalian epidermis and from the skin shed by frogs, to find that while the latter withstood the ni">t vigorous and long-continued action of gastric juice, the epi- dermal cells from the fish were wholly decomposed by the same treatment, so that no trace of them could be found. FTkidklbkrg and New Haven, IS*'.). REASONS WHY Physicians should Subscribe The New York Medical Journal, Edited by FRANK P. FOSTF.R, M.D., Published by D. APPLETON & CO., 1, 3, & 5 Eond St. i. BECAUSE : It is the LEADING JOURNAL of America, and contains more reading-matter than any other journal of its class. 2. BECAUSE : It is the exponent of the most advanced scientific medical thought. 3. BECAUSE : Its contributors are among the most learned medi- cal men of this country. 4. BECAUSE: Its "Original Articles" are the results of sci- entific observation and research, and are of infinite practical value to the general practitioner. 5. BECAUSE : The " Reports on the Progress of Medicine," which are published from time to time, contain the most recent discoveries in the various departments of medicine, and are written by practitioners especially qualified for the purpose. 6. BECAUSE : The column devoted in each number to "Thera- peutical Notes " contains a resume of the practical application of the most recent therapeutic novelties. 7. BECAUSE : The Society Proceedings, of which each number contains one or more, are reports of the practical experience of prominent physicians who thus give to the profession the results of certain modes of treatment in given cases. 8 BECAUSE : The Editorial Columns are controlled only by the desire to promote the welfare, honor, and advancement of th; science of medicine, as viewed from a standpoint looking to the best interests of the profession. 9. BECAUSE : Nothing is admitted to its columns that has not some bearing on medicine, or is not possessed of some practical value. 10. BECAUSE : It is published solely in the interests of medicine, and for the upholding of the elevated position occupied by the profession of America. frabacription Prioo, $6.00 per Annum. Volumes begin in January uiid July. COLUMBIA UNIVERSITY LIBRARIES This book is due on the date indicated below, or at the expiration of a definite period after the date of borrowing, as provided by the rules of the Library or by special arrange- ment with the Librarian in charge. DATE BORROWED DATE DUE DATE BORROWED DATE DUE C28(ll4l)M100 QM575 K95 Kuhne On neurokeratin. . ?i RE