ee as ie ih 
wu Ruth Mt eerie : 
vii 0 4 
+ =. > : H ; 
x itis : 
an 
. ie 
yt 
nat ti 
4 


i 


i 


4 
4 


i 


ith, 
* 
Varey 

‘| ho TH 
+ 
ith 


an 
. 
cy 


inte heaat 
+058 3ti8 
is rhe ve 


3° 
sare) 
a) 


ts. 
MBLC OH 
iy 
¥ 
iFoay : 
i thet e 
i eke 


tei9 
#4 
Aare 

bY , 7) 

tae bem tcoe) 


ret ebra shan 


te 
+iae 
: 


f 
eae 
ass 


rie TOUR NA L 


OF 


Comparative Neurology | 


A OUARTERLY PERIODICAL DEVOTED TO THE 


Comparative Study of the Nervous System. 


EDITED BY 
C. lL. Herrick, Socorro, NEw MExIco. 


aSSOCIATED WITH 
OLIVER S. SyRONG, COLUMBIA UNIVERSITY, 
C. JupDsoN HERRICK, DENISON UNIVERSITY. 


AND WITH THE COUtLABORATION OF 


LEWELLYs F. BARKER, M.B., University of Chicago and Rush Medical College 
Dr. FRANK J. COLE, University of Liverpool; HENRY H. DONALDsON, Ph.D., 
University of Chicago; PROFESSOR LUDWIG EDINGER, Frankfurt a-M.; 
PROFESSOR A. VAN GEHUCHTEN, Untversitté de Louvain; C F. Hopce, Ph.D., 
Clark University; G CARL HuBER, M.D., University of Michigan ; B. F. 
KINGSBURY, Ph.D., Cornell University and the New York State Vetert- 
nary College; FREDERIC S. LEE, Ph.D., Columbia University; ADOLF 
MEYER, M.D., Pathological Instetute, New York; A. D. 
MorrILL, M.S., AHamzlton College; G. H. PARKER, S.D., 
Harvard University. 


VOLUME XIII, 1903. 


DENISON UNIVERSITY, GRANVILLE, OHIO, U.S. A. 


| PU le or Ps thd 
« ioe Vala 
aw ¢ Pare aa! Lei aS. 
in ‘ai 
AO A a GS 
Ay. aS fia « 
‘ay Ce ta! Nene 
Ay " 9 ts oy, 
f ‘ | 
¥ \~ ’ qt Fue 
CURR Tove: Sauer 
¢ y, nl Pt: 
x »? é. 
if i 
J 
‘i 
¥ 
i 
4 
‘ 
, 
» 
i 
| 
. 
ibite 
ee 
Pre. 
Ve 
. 
aptt 
5 
i 
' 
' 
A 
ul 
* 
‘ 


THE 


JournaL oF Comparative NeEuro.oey. 


ContTENTS oF VOLUME XIII, 1903. 


No. J, APRIL, 1903. 
Pages 1—52. i—xx. Plates I—II. 
CONTRIBUTED ARTICLES. 


The Fore-Brain of Macacus. By Wm. WoLFE LEsEM, M. A. ( Columédzta 
Cuivezsity.\ an Wwithprlatesmy andi aes ssa ne eee eee ee 
Brain- Weights of Animals with Special Reference to the Weight of the 
Brain in the Macaque Monkey. By EDWARD ANTHONY SPITZKA, 
M.D. (From the Anatomical Laboratory, Columbia University.) 
A Description of Charts Showing the Areas of the Cross-Sections of the 
Human Spinal Cord at the Level of Each Spinal Nerve, By 
Henry H. DonALpson and DaAvip J. Davis. (rom the Neuro- 
logical Laboratory of the University of Chicago.) With Chart I__-- 
The Brain of the Archzoceti. By G. ELtior SmMirH, M. A., M. D., 
Fellow of St. Johns College, Cambridge; Professor of Anatomy, Egy- 
tian School of Medicine, Cairo. With four figures in the text_____ 


LITERARY NOTICES 


Functional Changes in the Dendrites of Cortical Neurones-._.-.. 
The Morphological Position of the Chorda Tympani in Reptiles__ 
Mendel and Jacobsohn’s Jahresbericht: Fifth Issue_....--_____- 
Nervoustoystenion Myxines- i. oe eb eee oe 
Waste and the Fifth Nerve 2—-=222—--<- ee eS SR Co eee 
be) Phyloceny of the Palliam, 22/2000 2 ooo ceo ee 
Obsessions and Psychasthenia 
MeMunrichy sem bryOlo gy) aaa 2 ee Se ere yee ye 
MotomNerveslenminiiiniimsectSeas=--s-s2eo ae oeee ere n eae 
The Comparative Anatomy of the Brains of Lemurs and Other 

Mammals 


9 


41 


No. 2, JUNE, 1903. 
Pages 53—156, xxi—xxviii. Plates II—IV. 
CONTRIBUTED ARTICLES. 


An Enumeration of the Medullated Nerve Fibers in the Dorsal Roots of 
the Spinal Nerves of Man. By CHARLES INGBERT. (From the 
Neurological Laboratory of the University of Chicago.) With thirty- 


GOMNCUTESS 22 ae oe ES SL oe Bo a) Se ee 53 
On the Phylogeny and Morphological Position of the Terminal Buds of 
Fishes. By C. Jupson Herrick. (Studies from the Neurological 
Laboratory of Dentson Universtty, No. XViD) eee aoa 121 
On the Nature of the Pericellular Network of Nerve Cells. By SHINKISHI 
Hatat. (From the Neurological Laboratory of the University of 
Chicago.) With ‘Plate TIl..- -2.2).- 2-2 eee 139 
The Neurokeratin in the Medullary Sheaths of the Beveneral Nerves of 
Mammals. By SHINKISHI HaTal. (from the Neurological Labor- 
atory of the University of Chicago.) With Plate IV_.----. ee 149 
LITERARY NOTICES. 
Hearing and Allied Senses in) Pishes) 22222562 eae eee XX1 
Taste Fibers and their Independence of the Trigeminus_-___ -_-- XXiil 
DevKursac’s Psychiatry-2.--.85=. Skee eee ee ee XXIV 
hesEvolution of Man and his Mind) 2222) --5-— === =e XXV 
The Brain and Nerves of the Anamnia+-2-222222> 5222252 XXV 
he Dorsal Spino-cerebellar Dract=-= =) == ee XXxvi 
The Optic Chiasma in Symmetrical and Asymmetrical Teleosts-_  xxvi 
Brain, Weights of Eminent Men: <2-22222 25. 22s cs eens eee xxvii 
dhe Lateral Sensory System’ of whe Belsi222=222 2-2 oa eee XXvili 
No, 3, OCTOBER, 1903. 
Pages 157—258. Plates V—VII. 
CONTRIBUTED ARTICLES. 
The Neurofibrillar Structures in the Ganglia of the Leech and Crayfish 
with Especial Reference to the Neurone Theory. By C. W. PREN- 
Tiss. Parker Fellow in Zoilogy, Harvard University. With Plates 
Wrand: Wiles.) cect cos ae et eee ee eee ee 157 
On the Increase in the Number of Medullated Maree Fibers in the Ven- 
tral Roots of the Spinal Nerves of the Growing White Rat. By 
SHINKISHI Hatal. (from the Neurological Laboratory of the Uni- 
DERSELIV OP A CHICOL OL) 3 ax Oe ta See ene eee ne en 77 


On the Medullated Nerve Fibers Crossing the Site of Lesions in the 
Brain of the White Rat. By S. WALTER RANSON. Jnstructor in 
Anatomy, Marion-Sims-Beaumont Medical School, St. Louis Univer= 
sity. (From the Neurological Laboratory of the University of Chicago.) 


NVainhp la tem Vallis sees Soka ieee eee oe ae ee eS 185 

On the Density of the Cutaneous Innervation in Man. By CHARLEs E. 
INGBERT. (From the Neurological Laboratory of the University of 
CHICAS GS a eee nee ae oe Se ee a pe a hs Vat is Sa 209 

On a Law Determining the Number of Meguilared Nerve Fibers inner 
vating the Thigh, Shank and Foot of the Frog—Rana virescens. 

By Henry H. DonALpson. (from the Neurological Laboratory of 
CREM RICET SIV OFS ORICA POs) eae nye ee einem eee See 223 
No. 4, DECEMBER, 1903. 
Pages 259—335, xxix—xliv. 
CONTRIBUTED ARTICLES. 

The Rate of the Nervous Impulse in the Ventral Nerve-Cord of Certain 
Worms. By O, P. JENKINS and A. J. CARLSON. (From the Hopkins’ 
Seaside Laboratory and the Physiological Laboratory of Leland Stan- 
ord a) tC river sii.) With tourteensucunest os] s— ee see so ace ee 259 

Notes on the Technique of Weigert’s Method tor Staining Medullated 
Nerve Fibers. By OLIVER 5S. STRONG, Columbia University, New 
Vor eed ere ee Nae me ene a ae ates ek oes aa Se 291 

The Doctrine of Nerve Components and Some of Its Applications. 
Byee.7 \UDSONW EER RICK? Sooo 22 oe eee, Oe eee 301 

Colnmella Auris and Nervus Facialis in the Urodela. By B. F. 
[SID VEIS Eta eM tM ee ENGL ENROL EA NOT Be RENCE NINE ABCD 313 

BditonaleAnnouncement sins. 22s. 2 See Sows sos oe eee ee 335 

LITERARY NOTICES. 
ihe Relations of) Biology and! Psychology =—2—- —2= - 2 Seer XX1X 
ANG: TESS ACLNONO TE COVE VeNC LNCS) ON ae Oe Sa Oe oore XxXxXill 
he svore-brain ofthe yBird2 0.220) Aes ee See ee ee AXE XIE 
The Optic Chiasma and the Post-optic Commissure.__.---------- xlii 
Peripheral Nerve Endings in Amphioxus________._------------ xlill 
On the Lobus Impar of the Brain of Cyprinoid Fishes.___-____- xliv 


VotumMeE XIII. 1903. NUMBER I 


THE 
JournaL or Comparative Neurovocy. 


4 


THE FORE-BRAIN OF MACACUS. 


By Wm. Wo tre Lesem, M. A. 
With Plates I and II. 


The present article is the result of investigations upon the 
simian brain pursued by the writer during the winter of Ig00 
in the anatomical laboratory of Columbia University. A 
careful search of the bibliography of the simian brain failed to 
show any article dealing with the species in question. Here 
and there the writer came across rather poor drawings of the 
macaque brain but nowhere did he meet with any description 
thereof except in Flatau and Jacobsohn’s Comparative Anat- 
omy of the Central Nervous System of Mammals. In the 
Anatomical Museum of Columbia University the writer has 
had access to a sufficiently large number of macaque brains to 
render his observations fairly accurate. 

Compared with the dog, Macacus presents many striking 
advances. A higher type of gyri and sulci, a well developed 
occipital lobe with a posterior cornu, a calcar avis, a prominent 
forceps major, and an enormously developed temporo-sphenoti- 
dal lobe are some of the points which will be discussed in the 
ensuing pages. 

Whereas in the dog the ventral aspect of the encephalic 
segments lies almost entirely in the same plane, the brain of 
Macacus presents that convexity of the pons Varolii so marked 
in man. The pons of Macacus differs moreover from that of 
the carnivores in being actually larger than the medulla. In 
the carnivores the crura cerebri are plainly visible upon the base 
of the brain as two large bundles of fibres which extend from 
the pons and diverge to enter the cerebral hemispheres opposite 


’ 


2 JoURNAL OF COMPARATIVE NEUROLOGY. 


the optic chiasma. The crura of Macacus are not visible on 
the base of the brain. They are entirely hidden from view by 
the well-developed temporo-sphenoidal and the occipital lobes. 
In contradistinction to dog and man Macacus presents but a 
single corpus albicans, which occupies a mesial position. We 
naturally expect to find a well-developed olfactory apparatus in 
dogs; and such is the case, for the olfactory bulb projects be- 
yond the hemispheres. The olfactory lobes, the bulbs and the 
olfactory tracts of Macacus are greatly atrophied. Compared 
with the dog the orbital surface of the frontal lobe has likewise 
retrograded. 


The Hemispheres. 


The growth backward of the body of the lateral ventricle 
in Macacus results in the formation of a posterior cornu, and 
gives us an additional lobe, the occipital. The occipital lobe is 
merely a differentiation of the posterior part of the parietal lobe. 
In Macacus it is very extensive, comprising about a third of 
the entire hemisphere. In man the occipital lobe comprises 
less than a sixth of the secondary fore-brain. 

The hemispheres of Macacus are in contact with one 
another throughout their entire extent; even the apical portion 
of the frontal lobes are contiguous. In the dog the frontal 
apices are slightly separated, and the rounded posterior extrem- 
ities of the parietal lobes diverge considerably to permit of the 
reception of the worm of the cerebellum between them. The 
entire hemisphere of Macacus is slightly curved from before 
backwards. This is most marked in the frontal lobe, where 
the apex curves sharply downwards in a hook-like manner. 
This curve is due to the non-development of the orbital sur- 
face. Externally the frontal lobes are flattened. The parietal 
and the occipital lobes are curved only slightly. 

The arrangement of the fissures and the convolutions of 
Macacus closely resembles the condition met with in the human 
foetus. The numerous annectent gyri of the adult human 
brain renders it difficult to homologize the gyri of the adult 
brain with those of Macacus. The Sylvian, the Rolandic, the 


LresEM, Fore-Brain of Macacus. 3 


inferior occipital, the par-occipital, the calloso-marginal, and 
the calcarine fissures divide the hemispheres of Macacus into 
frontal, parietal, occipital, and limbic lobes. A brief descrip- 
tion of these fissures is therefore desirable at this juncture. 

The Sylvian fissure is oblique in direction. In man it is 
horizontal, in carnivores almost vertical. It starts at the base 
of the brain, separating the orbital surface of the frontal from 
the temporal lobe. At first it proceeds laterally, then ob- 
liquely upwards and backwards, and finally terminates by join- 
ing the parallel fissure. In one specimen of Macacus nemes- 
trinus under observation, the Sylvian fissure did not join the 
temporo-sphenoidal. Owing to the non-development of the 
orbital surface of the frontal lobe, the sylvian fissure in Macacus 
rhesus has no anterior limb. In Macacus nemestrinus, how- 
ever, there exists a small rudimentary fissure occupying the 
position of the anterior Sylvian limb, but not continuous with 
the Sylvian sulcus. 

The fissure of Rolando is very similar in its course to that 
of man. It runs from just below the great longitudinal fissure 
and curves sharply forwards and downwards. Before its termi- 
nation, just above the mesial portion of the Sylvian fissure, the 
fissure of Rolando curves slightly backwards. 

The par-occipital fissure runs sinuously from the great 
longitudinal fissure almost to the inferior margin of the hemis- 
phere. Externally it is hidden from view by the growth for- 
ward of the occipital lobe. This growth forward of the occi- 
pital lobe results in the formation of a fissure, the ‘‘Affen- 
spalte.’’ This fissure, called by some the external occipital 
sulcus, is peculiar to Quadrumana. Upon drawing aside the 
‘“‘Affenspalte” we find lying in its floor two annectent gyri 
belonging to the par-occipital fissure. The par-occipital fissure 
is joined by the parallel and the intraparietal sulci. 

The inferior occipital fissure extends from the posterior 
margin of the occipital lobe almost to the parallel fissure. It 
is curved forwards and upwards, and passes just below the 
‘‘Affenspalte.”’ 

The calcarine fissure lies upon the mesial aspect of the 


4 JOURNAL OF COMPARATIVE NEUROLOGY. 


occipital lobe. It separates the temporal lobe from the occi- 
pital. At its origin it consists of two parts, an upper limb 
which arises from the superior margin of the occipital lobe and 
a shorter, lower limb which commences a little below the centre 
of the occipital lobe. These two limbs unite midway between 
the upper and the lower margins of the cerebrum. The calca- 
rine fissure then runs at first downwards and forwards to the 
base of the hemisphere. It then passes forwards and slightly 
upwards, and terminates in the dentate fissure. 

The calloso-marginal sulcus extends from the middle of the 
great longitudinal fissure almost to the apex of the frontal lobe. 
At first it curves downwards, then courses directly forwards, 
and finally curves upwards to terminate in the great longitudinal 
fissure. Before curving upwards it gives off a short inferior 
limb. 


The Frontal Lobe. 


The frontal lobe presents three surfaces, external, internal 
and orbital. The external surface is bounded inferiorly by the 
Sylvian fissure and posteriorly by Rolando. The external 
surface presents two fissures, the precentral and the horizontal. 
The precentral fissure is vertical in direction, and lies midway 
between the apex of the lobe and the fissure of Rolando. 
Superiorly it terminates by dividing into an anterior and a 
posterior limb. 

The horizontal fissure lies a little posterior to the apex of 
the frontal lobe. It runs at right angles to the precentral sul- 
cus. Posteriorly it curves slightly upwards. 

The precentral and the horizontal fissures divide the 
frontal lobe into three convolutions. Of these the most pos- 
terior one corresponds exactly to the ascending frontal gyrus of 
man being bounded anteriorly by the precentral sulcus, inferi- 
orly by Sylvius and posteriorly by Rolando. Just what the 
remaining two gyri are is difficult to say. BiscHorF declares 
that in the gorilla and the orang Broca’s convolution is but ill- 
developed ; and according to Panscu the third frontal convolu- 
tion is absent in all other apes. If Panscu be correct in his 


Lrsem, Fore-Brain of Macacus. 5 


theory we may homologize the upper frontal gyrus of Macacus 
with the first frontal of man, and the remaining lower gyrus 
with the second frontal. 

The mesial aspect of the frontal lobe presents the marginal 
gyrus enclosed by the calloso-marginal fissure. The gyrus 
fornicatus is continuous with that portion of the frontal lobe 
which in man is known as the precuneus. 

We have already noted the concave non-developed orbital 
surface of the frontal lobe. 


The Parietal Lobe. 


The parietal lobe has two surfaces; an external one 
bounded anteriorly by Rolando, inferiorly by Sylvius, and pos- 
teriorly by the ‘‘Affenspalte,’’ and a mesial surface which is 
continuous with the gyrus fornicatus. The external surface 
presents the intra-parietal fissure. This sulcus lies between the 
fissure of Rolando and the external occipital sulcus. ~It starts 
a little posterior to the fissure of Rolando, and passes upwards 
and backwards. Then it curves upon itself, and terminates in 
the par-occipital sulcus. The intra-parietal fissure was origi- 
nally composed of three distinct fissures, a superior, an inferior, 
and a horizontal parietal sulcus. This condition is still evident 
in the adult human brain where the three parts of the intra- 
parietal sulcus are separated from one another by numerous 
annectent gyri. 

The external surface of the parietal lobe presents two gyri, 
a superior and an inferior parietal gyrus. The superior parietal 
gyrus represents the ascending parietal and the superior parietal 
convolutions of man. The inferior parietal gyrus may be 
divided into supramarginal and angular convolutions. The 
supramarginal is bounded anteriorly by the intra-parietal sulcus, 
posteriorly by the parallel fissure, and inferiorly by the fissure 
of Sylvius. The angular is bounded in front by the temporo- 
sphenoidal sulcus, below by the inferior occipital, and behind 
by the external occipital. 


6 JOURNAL OF COMPARATIVE NEUROLOGY. 
Temporo-Sphenotdal Lobe. 


This lobe is very highly developed in Macacus. Its ex- 
ternal surface presents the parallel fissure which is the most 
extensive sulcus of the cerebrum. It starts near the apex of 
the temporal lobe, and extends almost to the great longitudinal 
fissure; and there it joins the parieto-occipital fissure. Inferi- 
orly it curves slightly forward; while superiorly it curves 
slightly backwards. The parallel fissure divides the temporal 
lobe into an upper and a lower gyrus. The lower gyrus is 
continued on to the base of the lobe. The superior convolu- 
tion lies below the supramarginal gyrus. It is bounded anteri- 
orly and above by the Sylvian fissure, and posteriorly and be- 
low by the parallel fissure. The base of the temporo-sphen- 
oidal lobe is divided into two convolutions by the collateral 
fissure. This sulcus runs almost parallel with the inferior mar- 
gin of the temporal lobe. It extends from within a short dis- 
tance of the posterior temporal pole almost to the apex of the 
lobe. The collateral fissure forms the lower boundary of the 
hippocampal gyrus. 


The Occtpital Lobe. 


As in man, the occipital lobe of Macacus presents an 
external, an internal and a basal surface. Externally with the 
exception of the superior occipital sulcus the occipital lobe is 
entirely smooth. The superior occipital fissure runs at right 
angles to the external occipital fissure. Externally the occi- 
pital lobe is separated from: the parietal by the par-occipital 
fissure. The inferior occipital fissure separates the occipital 
from the temporal lobe. The mesial surface of the occipital 
lobe presents the cuneus, bounded in front by the parieto- © 
occipital fissure and behind by the calcarine. 


The Limbic Lobe. 


The limbic lobe is bounded above by the calloso-marginal. 
fissure, posteriorly by the rudimentary post-limbic sulcus, 
and inferiorly by the collateral fissure. This lobe includes the 
hippocampal and the fornicate gyri. The gyrus fornicatus 


Lesem, Fore-Brain of Macacus. 7 


extends from the genu of the corpus callosum to the lower 
border of the splenium. Here it becomes the hippocampal 
gyrus. The latter is limited above by the collateral sulcus. In 
Macacus as in man the hippocampal gyrus forms a convolution 
of the temporal lobe, and is fused with the rest of hemisphere. 
In the brain of the sheep this is not the case; the hippocampus 
has no connection whatsoever with the exterior of the brain. 
In Quain’s ‘‘Anatomy of the Human Brain’ BEEvor is quoted 
as saying that the hippocampus major of apes receives no 
fibres whatsoever from the lyra of the fornix. To determine 
this point I carefully dissected two brains of Macacus and 
found in both a distinct band of fibres running from the lyra of 
the fornix to the hippocampus major, 

The island of Reil in Macacus consists of two convolu- 
tions derived from the orbital surface of the frontal lobe. As 
in man the insula is overlapped by the operculum, and is thus 
invisible on the exterior of the undissected brain. In carni- 
vores owing to the non-development of a large temporo- 
sphenoidal lobe, no insula exists. 


The Lateral Ventricles. 


The lateral ventricles of Macacus are especially interesting 
owing to the enormous extent to which the posterior cornu is 
developed. This development is not uniformly attained by 
man. Numerous cases are on record where the posterior 
cornu has been of small size or rudimentary. In the past year 
I have seen two brains, one that of a child two years old, the 
other an adult brain of thirty years, in which the posterior 
cornu of the left side measured 1.5 cm. and 2 cm. respectively. 
The right posterior cornu was well-developed in both cases. 
The posterior horn of Macacus presents a well-developed bulb, 
but an ill-developed calcar avis. The anterior and the descend- 
ing cornua present no features markedly different from those 
of man. 

The corpus striatum of Macacus is remarkable in that the 
lenticular nucleus seems to exceed the caudate in size. As in 
man the lenticular nucleus consists of three parts, an outer 


8 JOURNAL OF COMPARATIVE NEUROLOGY. 


portion known as the putamen, and two inner divisions called 
the globus pallidus. In carnivores no such division exists. 

The corpus callosum of Macacus is very short. It pre- 
sents, however, a well-developed forceps major and a forceps 
minor. The nerves of Lancisi are also plainly discernible. 
These cannot be differentiated in the dog. The splenium of 
the corpus callosum sends off a thick bundle of fibres which 
fuses with the hippocampus of both sides and serves as a com- 
missure between them. 

The anterior commissure is more extensive than in carni- 
vores. It runs into the temporal lobe. Its termination can be 
ascertained only by microscopic methods. 


The Third Ventricle. 


The cavity of the third ventricle is extremely small and 
narrow. This is due to the fusion across the median line of 
the large optic thalami, which are continuous with one another 
throughout their length except inferiorly. This fusion across 
the median line represents the middle commissure of man. 

From this brief macroscopic study of the fore brain of 
Macacus we see that the simian brain closely resembles that of 
man; for Macacus seems to present most of the conditions 
existing in man, the sole difference in most instances being one 
of degree. 


DESCRIPTION OF PLATES. 
PLATE I. 


fig. z. The external surface of the left hemisphere of Macacus rhesus. 
fig. 2. Mesial surface of left hemisphere of Macacus rhesus. 


PLATE II. 


fig. 7. Mesial surface of the left hemisphere of Macacus nemestrinus. 
fig. 2. External surface of cerebrum of Macacus nemestrinus. Part of 
the occipital lobe has been removed to show the annectent gyri. 


JOURNAL OF COMPARATIVE NEUROLOGY. VOL. XIII. PEATE. 


Rolando 
Horizon- 
tal fissure Affen- 
spalte 
Precentral ee 
; nferior 
Sudineay Occipital 
Parallel 
Sylvius fissure 
Occipital 
Calloso- 
marginal 
Fornix 
Frontal 
Pons Lobe 


JOURNAL OF COMPARATIVE NEUROLOGY. VoL. XIll. 


Precuneus 


Occipital 
Sulcus 


Calcarine 
fissure 


Ascending 
Frontal 


Inf. trans- 
verse 
fissure 


Second 


Frontal 


PLATE Il- 


Calloso- 
marginal 


Annectent 
gyri 


Occipital 
Lobe 


Temporal 
Lobe 


My 
7 4 
A 


BRAIN-WEIGHTS OF ANIMALS WITH SPECIAL REF- 
ERENCE TO THE WEIGHT OF THE BRAIN IN 
THE MACAQUE MONKEY. 


By Epwarp ANTHONY SpitTzKa, M. D., 
Alumni Association Fellow in Anatomy, Columbia University. 


(From the Anatomical Laboratory, Columbia.) 


The accumulation of considerable material for morphologi- 
cal study during a period of over ten years has furnished a 
series of brain-weights whose publication as a contribution to 
. the subject from a comparative standpoint seems desirable.* 
All the animals whose brain-weights are here given were 
Mammals (204 in number), the great majority belonging to the 
Quadrumana, among them being 80 of the genus Macacus. Of 
the total number, 192 brains were weighed in the fresh state, 
the remaining 12 after the body had been injected with a zinc 
chlorid solution. The latter series is tabulated separately. 
The weights are expressed in grammes. In all cases the brain 
was severed from the spinal cord at the foramen magnum and 
weighed with its pialinvestment. In nearly all cases the body- 
weight was also recorded, giving the relative as well as the ab- 
solute brain-weight. In this connection it must be noted that 
many of the animals, particularly among the Quadrumana, were 
quite young, giving ratios of body and brain-weight not gener- 
ally applicable to the adult animal. Thus, among the gyren- 


1 The writer is indebted to Professor GEORGE S. HUNTINGTON for the 
privilege of compiling these data. 


10 JOURNAL OF COMPARATIVE NEUROLOGY. 


cephalic Mammals, we find such ratios as 1:17 for a young 
Chimpanzee, and 1:20 for a young Coati-mundi. Of course, 
such cases can not be taken as standards any more than they 
can in the human species, where the new-born babe, with a 
ratio of 1:8 has relatively five times as heavy a brain as an 
adult man. Furthermore it must be remembered that in the 
accession of such material as is comprised in this collection, 
most animals arrive in a poorly-nourished, half-starved condi- 
tion, or they have died as the result of wasting diseases; in. 
either case losing considerable body-weight and materially de- 
tracting from the value of any computations of any brain- and. 
body-weight ratios. It is fortunate that the brain attains to: 
nearly its largest size so early in life, and that its weight is so 
slightly affected by starvation and disease, and it is for this 
reason that the writer ventures to subject the large series of 
brain-weights of the Macaque monkeys to a special analysis. 
However, since the body-weight alone as a criterion or as a 
means of comparison is inadequate, it is our purpose hereafter 
to record as well the bodily dimensions, such as the length from: 
vertex to heel, or to the root of the tail, and particularly mea- 
surements of the head. 

The list of brain-weights, with the sex, body-weight, and 
ratio (the brain-weight is considered as equivalent to 1), 
follows : 


Spitzka, Brain-Weights of Animals. 
TABLE I. 


FRESH BRAIN WEIGHTS. 


— | SSeS 


1|Troglodytes niger, juv. 
2 sé “é “es 
Semnopithecus entellus, 


1,Cercopithecus callitrichus, 3159 64 49 
2 «6 66 2898 37 51 
3 66 «6 1679 64 26 
& % 845 59 T5 
66 66 2 140 54 40 
Cercopithecus ne-s=ss=—> 1502 62 24 
1/Cercopithecus mona, 3001 67 45 


2 se 6 
Cercopithecus griseo-viridis, 


é 
2 
ef 
3 
é 
2 
2 
a 
é 
3 
1|Chlorocebus sabaeus, juv. fe) 1036 50 21 
2 6 66 66 Rae 1480 71 21 
Chlorocebus cynosurus, ay 3880 68.5 57 
1|Cercocebus fuliginosus, eS eee es 97) \Sazanceee 
2 oe “ce & 1353 gI 15 
6e “cc ce) 1398 105 13 
1/Macacus rhesus, ay 5190 93 56 
2 “ s S 2620 92.5 28 
3 v5 rs ae e535 92 17 
4 46 “ a 1250 go 14 
5 mi a o 1173 87 13-5 
6 66 66 a 2072 86 24 
7 us Gu a 1928 86 20 
8 se es zB 1551 82 19 
9 se oe é 1327 81.5 16 
10 5 Sy S| 3079 81 38 
1! OG se Ag 1350 80 17 
12 OY OC & 1672 78.5 21 
13 “6 “ec Be 912 78 9 3 
14 i a $| 1400 77 18 
15 is n é 744 77 9-5 
16 oe Ss ze 1275 76 17 
17 a6 “sé é 897 75 12 
18 a ; $| 1079 75 14 
19 ee ie S| 1106 74 15 
20 se UG a 1602 74 21.5 
ae cs 3} 1353 73 18 
22 “se oe & 895 WB 12 
23 66 a6 as I 105 73 15 
24 ae “6 & eats BS ct 72 Be rea se 
25 os oc g 1248 64 19 
26 od Oe Q 2025.5 98 21 
27 i ay 2 1325 95 14 
28 an ne ce) 1207 89 13.5 
29 of “ < 1905 92 21 
30 ss 6s 9 1900 86 22 
“ 
4 


JouRNAL OF COMPARATIVE NEUROLOGY. 


Ratio: 
Brain- W't 


No QUADRUMANA Sex|Body-W’t|Brain-W’t) =I! 
Grams. | Grams. 
33|Macacus rhesus, Q 1685 81 21 
34 OC ac Q 2045 80 25.5 
35 a es Q 2286 79 29 
36 ad ae 9 1747 78 22 
37 a a Q 1626 77 21 
38 ae a °) 1656 77 21 
39 Sy e 9 974 76 13 
ek Saas - 2 960 75:5 13 
4! ae os Q 1187 73 16 
42 a ‘§ 9 1106 72 15 
43 as “ 2 1760 72 24 
44 a S Q} 1942 72 27 
45 OC we 2 1388 72 19 
46 oy a Q| 1342 71 19 
47 ? a 2 877 70 12.5 
48 oC $ 2 1344 67 20 
49 < K O 2E75 67 18 
50 OC (Tail amputated.)| 9? 1880 61 31 
5! et : =) § Aga 87 20 
52 os ol —|cir 2000 87 23 
53 ne 7 iy mega 83 21 
54 a as — 2180 82 26 
55 os Os —| 1180 82 14 
56 4 a —| 2020 73 28 | 
57 es J — 1621 72 22 
1|Macacus cynomolgus, 2 1567 67 23 
2 ee ue a 2070 62 33 
3 ae ss 3B 2060 62 33 
4 “6 “6 3 918 59 16 
5 a oC juv.| ¢ 582 50 12 
6 ee ie juv.| g 420 44 9-5 
7 “ «s juv.| ¢ 540 44 12 
8 ag “ co) 1203 100 12 
9 a ae juv.| 9 919 62 15 
10 = Q| 1803 54 33 
11 a ts ce) 1497 53 28 
12 a a fe) 1142 52 22 
13 “ ve Q 1126 47 24 
14 as fs — 1231 62 20 
1;Macacus nemestrinus, ay 8500 128 66 
2 oe ae o 8610 122 7t 
3 os ne é 5610 119 47 
+ > ca g 4590 118 39 
5 ae a6 a he 113 pene ry SN 
6 as as g 6620 105 63 
7 ti ee ce) 2287 103 22 
8 a OC ce) 3580 100 36 
9 i i 2 3100 95 32 
10 “so “c fe} 2S ee 84 we eee 
1|/Macacus sinicus, a 1340 75 18 
2 fs s By 848 68 12.5 
3 * g 793 58 12 
4 “ ' as fe) Tie ees 70 ui secooeee 
5 “6 46 ae (een Leb. 67 fe 
1|Macacus pileatus, a 1190 60 20 


SpitzKa, Brain-Weights of Animals. 


QUADRUMANA 


| ee 


2)Macacus pileatus, 


3 aa oe 
4 6 “6 


1|Macacus speciosus, 
2 “6 6 
Macacus melanotus, 
1\Cynopithecus niger, 
2 46 *6 


juv 


3 
1|Cynocephalus babouin, 
2 oe 6 


Cynocephalus hamadryas, 
46 ae 


“a a6 


juv. 
juv. 


© CONF ANN PW WD Hint Ww 


1|Cynocephalus anubis, 
2 46 6 
Cynocephalus mormon, 
Cynocephalus sphinx, 
Cynocephalus leucophezus, 
Mycetes cavaya, 

Mycetes ursinus, 

Ateles beelzebub, 
Lagothrix humbcldti, 


1|Cebus capucinus, 
La} é 


juv. 


“6 “ce 
“6 “6 


“6 of 


Cebus capillatus, 


Cebus hypoleucus, 


WR eH DH Binet WD 


Cebus subcristatus, 

Cebus annellatus (Vertex 
root of tail 42 cm.) 

Cebus albifrons, 

Chrysothrixz sciureus, 

Nyctipithecus comm., 

Hapale penicillata, 


1\Jacchus vulgaris, 
“6 a6 


Ww wb 


Sex 


Ratio: 
Brain-Wt 
Body—W’t|Brain-Wt| =1 
Grams Grams 
2.1357 72 19 
2 1403 64 22 
2 1801 62 27 
2 1262 57-5 22 
gd} 5560 98 57 
2 942 81 I1.5 
é 1105 80 14 
é 6350 lag 58 
g 5490 107 51 
ES eee 96 |---- ---- 
o 3340 167 20 
rs 3175 148 21 
g 1545 109 14 
Q| 2235 135 16 
2 2414 128 19 
o| 10230 199 51-5 
2 3332 153 22 
2 1678 136 12.5 
2 1692 114 15 
2 1572 U2 14 
Q 1490 108 14 
9} 3581 104 34 
od eee HO} + jee yeee 
—| 1100 96 11.5 
2} 5000 152 33 
Sees 167): |jsoaseeeee 
2 1292 103 12.5 
Q 2481 135 18 
2 1718 124 14 
3 826 45 19 
Q}| 1025 19 54 
2 1870 © 97-5 20 
— 4850 112 43 
Gi 2030 79 26 
oS 1083 76 14 
a 1812 Jo 26 
cg 111g 58 19 
2 1223 68 18 
é 1370 69 20 
_— 1660 77 21.5 
maf os O75) plese eS 
dll 1672 59 11.5 
co 504 50 10 
g 585 54 I! 
ry 902 71 13 
to|(without int’s- 
g |tin’s) 1490 69 22 
dite i675 58 12 
& 2410 22 109 
eee a AYES eee eee 
Q 206 8 26: 
Bs 320 8 40 
ry 204 7 29 
Q| 270 9 30 
298 8.5 35 


13 


14 JourNAL OF COMPARATIVE NEUROLOGY. 


Ratio: 
Brain-Wt 

No QUADRUMANA Sex|Body-W’t} Brain-Wt| =I! 
ey Grams. | Grams. 

5|Jacchus vulgaris, _ 204 9 23 

6 “ “ — 150 Yi 21 

7 “ “ nae 207 7 27 

8 “ “a = 105 7 15 

Midas ursulus (rufimanus) 9 361 24 15 

Lemur bruneus, Pa 1505 26 58 

Nycticebus tardigradus, By 612 13 51 

Carnivora. 
1/Ursus americanus, g| 25990 248 105 
2 “6 a6 & pete. AO OA 192 eos 
1|Nasua rufa, juv.| gf 719 35 20 
2 “6 “6 juv. a Ae ea es 30 pa 
3 “é ac 2 3200 41 78 
Alte a? as ; ce) 2165 32 66 
Gi a juv.| 9 1127 29 39 

Lutra vulgaris, gt 2215 39 57 

Bassaricus astuta, ae 842 19 44 
1/Vulpes fulvus, juv. Be 3070 49-5 61 
ae “ juv-| Q] 3458 53 65 

Felis tigrina, Q 1989 63 31 

Zalophus californ. (Gillespie) | —]--------- 335 = «|--------- 

Rodentia. 

Arctomys monax, 2 3051 II 277 
1|Dasyprocta agouti, >) Paeenne ee 10) lene 
2 . f z 1803 19 95 

6 ‘6 ‘2 ae 2935 21 133 
1|Cynomys ludovicianus, é 392 7-5 52 
2 “a “ fe) 504 7 72 

Ungulata 

Equus caballus, o) Ee ae 51925, |= seneee 

Dicotyles tajacu, 9 7930 74 107 

Auchenia glama, 2 (eee 222). ||-=asseeee 

Edentata. 
\Myrmecophaga jubata, | 2 | 18940 84 225 

Tamandua bivittata, 2 930 | 21 | 44 

Marsupia lia. 
Didelphis virginiana, = ee ee | ; Mb osc 
{Petrogale xanthopus, | 25 2441 23 | 106 


SpitzKa, Brain-Weights of Animals. 15 


The following list contains the fresh weight of the body 
but not of the brain, this organ having been removed some 


time after a zinc chlorid solution had been injected into the 
body : 


Zn Cl2 
Sex| Body-W’t}Brain-W’t 
Macacus rhesus, 1866 97-5 

“ as 2069 86 
“6 “6 1490 85 
‘es 66 1640 80 
«6 ‘6 805.5 78 
46 66 978 69 
Cercocebus albigena, 1991 76 
Cercocebus fuliginosus, 2561 94 
Gercopithecusreallitrichusys) 130), econ eeee 65 
“cc “cc 1629 75 
Cebus albifrons, 1640 80 
Cebus capucinus, 1409 69 


ANALYSIS OF 80 FRESH BRAIN-WEIGHTS OF THE GENUS MACACUS. 


Macacus rhesus. Among the 57 specimens there are 25 
of males, the same number of females, and 7 of which the sex 
is not recorded. The body-weight is wanting in the case of 
one of the males. Twenty of the males weighed over 1000 
grammes each, and 22 of the the females. Only one of these, 
a male weighing 5190 gms., with a brain-weight of 93 gms., 
can be considered a full-grown adult; the ratio is 1:56 and is 
doubtless much nearer the true adult ratio than any of the 
others in the list. The cases are tabulated according to sex, 
and average brain- and body-weight in the following table: 


TABU «iL 
Macacus Rhesus. 
Total Total Averages. 

No. of )Range of Body-| Body- Brain- Body- Brain- 
Sex. Cases Weight. Weights. |Weights. | Weight. | Weight. 
Males 20 (1079-5190 gms)| 34926.0 1615.5 1746.3 80.77 
1 4 |( 744- 912 “* )| 3448.0 303-0 861.0 75-75 
Females 22 (1106-2286 ‘* )| 35659.5 1725.0 1620.9 78.40 
“ 3 ( 877- 974 ‘* )| 2811.0 221.5 937.0 73.80 
(?) 7 j(1180-2180 ‘* )| 12485.0 566.0 1783.4 80.90 
Male I | Wee a see saan coaloneeen ees Te OM eee oe 72.00 


16 JOURNAL OF COMPARATIVE NEUROLOGY. 


Ratios of Body-weights and Brain-weights. 
(Brain-weight =1.) 


Total of 56 cases, I 20816: 
20 Males (1079-5190 gms.) I's 21562: 
22 Females (1106-2286 ‘‘ ) I? 20:672 


Macacus cynomolgus. There are 14 specimens of this 
species, 7 of males, 6 of females, and 1 of unknown sex. 
None of these, to judge by the body-weight, were full-grown 
animals. The highest brain-weight, 100 grammes, is that of a 
female weighing only 1203 gms., a case which must for the 
present be regarded as exceptional. The average weights 
range between 50 and 67 gms. for the larger animals of the 


series. 
TABLEMIE 
Total , Total Averages. 

No. of |Range of Body-| Body- Brain- Body- Brain- 
Sex. Cases Weight. Weights. | Weights.| Weight. | Weight. 
Males . 3 (1567-2070 gms){ 5697.0 191.0 1899.0 63.7 
a 4 ( 420- 918 ‘** )| 2460.0 197.0 615.0 49.2 
Females 5 (1126-1803 ‘* )} 6771.0 306.0 1354.2 61.2 
“6 1 ( 919) 919.0 62.0 919.0 62.0 
(?) I (1231) 1231.0 62.0 1231.0 62.0 


Ratios of Body-weights and Brain-weights. 
(Brain-weight =1.) 


Total of 14 cases, I 3/20,8 72 
3 Males (1567-2070) I 1 20,82: 
5 Females (1126-1803) I 3.220 as 


Macacus nemestrinus. In eight of the ten members of this 
species the body-weight had been noted (5 males; 3 females.) 
Two of the males, weighing respectively 8500 and 8610 gms. 
have brains weighing 128 and 122 gms., giving ratios of 1 : 66 
and 1: 71. 


TABLE IV. 
Total Total Averages. 
No. of |Range of Body-| Body Brain Body Brain 
Sex. Cases Weights. Weights. | Weights.| Weight | Weight. 
Males 5 (4590-8610 gms)} 33930.0 592.0 6786.0 118.4 


Females 3 (2287-3580 ‘* )' 8967.0 298.0 2989.0 96.0 


SpitzKA, Brain-Weights of Animals. 17 


Ratios of Body-weights and Brain-weights. 
(Brain-weight—1.) 


Total of 8 cases, Fo7540. 20; 
5 Males (4590-8610) ey Mee 
3 Females (2287-3580) erie fe 


Other Spectes of Macacus. Five specimens of M. sznicus, 
the same number of J. pileatus are all too young to furnish 
reliable ratios. The average brain-weight of MW. sénzcus (maxi- 
mum 75 gms.) is 67.6 gms. Of M. pileatus (maximum 72 
gms.) is 63.1 gms. 

An adult specimen of WM. sfeciosus (3) weighing 5560 
gms. has a brain-weight of 98 gms., giving a brain ratio of 
1:57. The brain of a young specimen (?) weighs 81 gms. 
A single specimen of J. melanotus has a brain-weight of 80 
gms. 

Judging from these records and allowing for disturbing 
factors, the following tabulation of the absolute and relative 
brain-weights, with their variations, may be here proposed. 
The sexual differences are not discussed at present, for the 
number of adult specimens is far too small for accurate analysis. 
Asa rule, however, the females seem to have a smaller brain- 
weight, both absolutely and relatively, although the reverse 
would appear to be true were the total of the tabulated cases to 
be alone considered. 

The list of ‘‘ Probable Averages” in Table V is only 
tentatively proposed, for the accession of a larger number of 
adult specimens may materially change certain of the figures. 


Bie Bei. 


PROBABLE AVERAGES OF BRAIN-WEIGHTS IN THE GENUS MACACUS. 


Probable Ay- 
erage Brain- 


Probable Av- 


Usual Range of erage Adult 


Brain-weight. weight. Ratio. 
Macacus rhesus, 70— 90 gms. 80 gms. Tc5s 
“«« cynomolgus, 50- 70 *§ Gouss I: 50? 
Ks nemestrinus, 95-120 ‘ LILO ee 170 
to sinieus; 65-75 <* 70; <é ? 
J pileatuss 60- 70 ‘< 65s I: 50? 


‘« _speciosus, 80-100 ‘ gone. I: 60 


Lee, Say 


A DESCRIPTION OF CHARTS SHOWING THE AREAS 
OF THE CROSS SECTIONS OF THE HUMAN 
SPINAL CORD AT THE LEVEL OF EACH SPINAL 
NERVE. 


By Henry H. Donatpson anp Davip J. Davis. 


(From the Neurological Laboratory of the University of Chicago.) 


A. CURVES SHOWING THE AREA OF THE CROSS-SECTION OF EACH 


SEGMENT OF THE MATURE SPINAL CORD. 
Introduction. 


The data which are presented in this paper were gathered 
for the purpose of preparing a curve based on the human 
spinal cord, with which to compare the areas of white and gray 
substance found in the cross sections of the spinal cords of 
other mammals. Reference to the literature shows that 
with the exception of the curve presented by KrausE and 
AGUERRE (1), which was published while this study was in pro- 
gress, the series of curves appearing in the text-books and used 
to show the areas of the gray and white substance at different 
levels of the cord, was first introduced by WoroscuitorF (2), 
- while that investigator was making a study of the conduction 
paths in the spinal cord of the rabbit. _WoRoSCHILOFF’s curves 
were based on measurements published by B. STILLING (3). 
The fact that the records which WoroscHILOFF chose as 
‘the basis for his curves were froma child of five years, and 
therefore from acord not completely developed, has been re- 
cently pointed out by several writers. That WoroscHILOFF 
‘should: have used these particular records of STILLING, instead of 
» taking the records for mature cords, published in the same vol- 


20 JouRNAL OF COMPARATIVE NEUROLOGY. 


ume, is explained by the fact that only in the case of the five- 
year old child are the areas for the separate funiculi given, and 
his interest was at that time directed to the funiculus lateralis. 
Since the white substance in the cord of the five-year old child 
is, both absolutely and proportionately, less than in the adult, the 
use of this series of curves to illustrate the gray and white sub- 
stance in the mature spinal cord is necessarily misleading, yet 
these curves are at present employed in the text-books, without 
any accompanying statement to show that they are based on 
the measurements from an immature cord. 

It is intended in this paper to present a chart which shall 
more accurately show the true relations between the gray and 
white matter as they appear in the adult, and thus shall replace 
the older charts now in use. In order to do this, not only 
should the measurements of the areas be those from the adult 
spinal cord, but there is another correction which applies to all 
the charts thus far published, including that of Krause and 
AGUERRE (1), and which consists in representing the segments 
of the cord in their true lengths. 


I. Representation of the Length of the Segments. 


Heretofore, in these charts, the abscissa has always been 
divided into 31 egual parts—each part representing the length 
of a segment of the spinal cord. Manifestly this will give an 
incorrect form to the curve, because the segments of the cord 
are really of unequal length. 

As any ordinate representing the area of a cross section 
applies strictly to the sum of half the distance from it to the or- 
dinates next above and next below the point at which it is 
erected, it follows that by multiplying the areas represented by 
any ordinate by the length of the cord to which it applies, we 
get an approximation of the volume of the segment. It is evi- 
dent, also, that the volume of a segment thus determined when 
the divisions of the abscissa are equal, would be different from 
that determined when the divisions of the abscissa represent 
the segments in their true length. To make a correct con- 
struction, it was therefore necessary to gather data on the 


DonaLpson AND Davis, Human Spinal Cord. re 


lengths of the segments of the spinal cord. The measurements 
of these lengths recorded by STILLING (3, p. 619), are insuffi- 
cient, having been made between the uppermost and lowest 
fila of each nerve—thus omitting some of the cord, where, as is 
conspicuously the case in the thoracic region, the line of roots 
is not continuous. A number of fresh observations were there- 
fore made. The measurements to determine the lengths of the 
segments in the adult human cord were made on one specimen 
(W)—preserved in normal size in 10 per cent. formalin;—on the 
two careful delineations (X, Y) published by Kany (4), and on 
the photolithographic chart (Z) of RipINGER (5). In the three 
plates the cord is depicted in natural size. The cords in the 
order named, are designated W, X, Y, Z. 

In the cord W, the condition of the specimen did not per- 
mit the measurement of the first four cervical segments. In 
cords X and Y, the first cervical segment could not be mea- 
sured on the dorsal aspect nor on the coccygeal segment at all. 
In cord Z, measurements on the dorsal aspect alone could be 
made, and even these could not be extended below the level of 
the 12th thoracic segment. 

To determine the length of a segment, the distance be- 
tween the uppermost fila of successive nerves was found, begin- 
ning with the uppermost filum of the first cervical nerve.' This 
was done both on the dorsal and ventral aspects of the cord. 
In making the measurements, the distance was marked off with 
a pair of spring compasses, and then this distance was measured 
on a metal scale to the nearest tenth of a millimeter. Each 
measurement has been separately entered in Table I. 


1This method of measurement credits to any segment the entire ‘‘inter- 
segmental space’’ which lies caudad to it. The method of LUDERITZ (6) was 
to credit to any segment one-half the length of the intersegmental spaces lying on 
either side of it. The difference in the results would be the amount by which, 
in any instance, the intersegmental space caudad to the part of the segment to 
which the roots were attached differed in length from the sum of half the inter- 
segmental spaces above and below the same parts. This difference in general 
would be small. 


22 JOURNAL OF COMPARATIVE NEUROLOGY. 


TABEE. 1: 


Showing in millimeters the lengths of the segments in the adult 
human spinal cord. 


Segment Cord W Cord X Cord Y Cord Z Average 
Aspect Dorsal Ventral|Dorsal Ventral|Dorsal Ventral Dorsal 

(1 5-9 8.2 7.6 7-23 
| 2 Holly i iPe8) 6.0 13.0 10.4 9-74 
"eta 12.2 11.0 | 12.0 13.1 13.8 12.40 
: | 4 GHP Seige iy nS 18.1 14.34 
& 15 TOSG, Lie 7 ish eMicey dl SyLfoy ste 13.9 12.35 
.S) | 6 14-3 140 13-7) 14-0) | ange ia 13.5 13.75 
7 T2254. 1226) 0522) OLS eenico 13-7 12.85 
L8 12:6) 91'5.7 15-6 3033006) e227, 12.1 13.20 
{1 122) a 146) ig s7a ate 9.3 13.4 12.80 
2 [2:90 1405 12:95) W523) eksese wake 16.1 14.20 
| 3 Hie7 t= Foe) 18.8 18.8 | 15.6 "16.4 16.6 17.30 
4 16.7. 20.2 21.0)  1953\7|| 24269200 24.3 20.90 
LOR is 19.8 19.9 20:3 20.5 | 23-6) 26.0 22.5 21.85 
2} 6 25°83". 26.6) °} 24.1! 22:21 arog eek 21.9 23.55 
8 7 287 28.8 1723), 20.7 2o:Aw zee 23.8 24.22 
ayiile ce 29:6))28:3 19:8) 1120.05 |)28.0 sezaeo 23.9 25.05 
9 22:8) 23.3 19:0, (22-44) 2458) 626.3 25-7 23-50 
| 10 22.7) 2564 19,7.) S20 sul O-Oreekee 26.2 22.50 
II 280 20.1 19:1") | 20.43) ) 2335 Qee 19.2 21.38 
er2 20.2 20.9; |(20.3,. F3.9 |/2h.50 19-5 19.60 
eal 21-3 23-0" | 1595) 0427) 10.3 akoe2 18.30 
g\2 R54 937 VO) 1955 i] is Omen mnO 12.90 
g13 13-2) 114-0 III Onja| MOS) ak25 11.80 
3 [3 YO2S0) LEO 10,7 728) || LOIQy Lies 10.64 
5 9.1 7.2 10.5 6.2 9.9 6.1 8.16 
I Vols 8.4 10.3 Bay 8.6 6.1 7.74 
S| 2 6.1 6.6 9:6, > “4 .5n\org.4 8.3 8.40 
943 6.6 4.4 8.4 4.8 8.3 9-9 7.07 
nN |14 6.1 6.6 5-4 4-7 8.5 8.6 6.65 
5 6.6 4.1 3-3 38 | 4.9 5.9 4:77 
41 2.5 2.50 


Coccygeal 


The last column in the Table (I), gives the average of all 
the measurements for each segment, and these average lengths 
were those used for the divisions of the abscissa in the accom- 
panying chart I. 

(a) Total Length of the Spinal Cord. The lengths of all 
the segments taken together, should, of course, equal the 
length of the entire cord. The average length obtained for the 
segments will therefore depend on the lengths of the cords. But 
before presenting the lengths of the cords which were here ex- 


LE EE 


DonaLpson AND Davis, Human Spinal Cord. 23 


amined, it will be best to state what is already known concern- 
ing the length of the human spinal cord. 

The correlation in development between the medulla 
spinalis and the columna vertebralis indicates that the longer col- 
umna vertebralis would contain the longer medulla spinalis. 

The observations of RAVENEL (7) on the length of the adult 
human spinal cord, show in 11 adult males, a range in length 
of 390-480 mm., with an average of 448 mm. In 11 adult fe- 
males, the range is from 370-460 mm., with an average of 413 
mm. _ These results plainly exhibit the greater average length 
of the medulla spinalis in the male. The measurements were 
made on fresh material from the level of the upper edge of the 
atlas to the lowest filum of the coccygeal nerve on the conus 
medullaris. 

The cords examined for the length of the segments in the 
present investigation give the following lengths: 


TABLE II. 
Cord Sex Length in mm. 
WwW Male 458 
Xx ? 403 
Y ? 453 
Z Male 448 
Average, 440.5 


It is thus seen that this average length lies between the av- 
erage length for the males and that for the females, though 
rather nearer the former, as determined by Ravenet (7). It 
concerns us here, however, merely to show that the average 
length obtained is a medium one—differences according to sex 
being disregarded. The average lengths of the segments deter- 
mined from the several cords (W, X, Y, Z) are presented in 
Table I. | When these average lengths are summed, they give 
441.6 mm., for the length of the cord, against 440.5 mm., as 
determined by direct measurements. The former number is 
the one employed in the construction of the chart. 

LUperiTz (6) is the only investigator who has made a de- 
tailed study of the length of the segments of the spinal cord. 

On comparing the results just given (Table I), with those 


24 _ JouRNAL OF COMPARATIVE NEUROLOGY. 


obtained by Liperirz (6), it appears that the averages of the 
sums of the segments in: his two men—33 and 37 years of age 
—show the total to be 450.5 mm., or about 2% more than the 
total sum in our case. This is a difterence which is well within 
normal limits, since the measurements of RAVENEL (7) show for 
the male cord a range in length from 390-480 mm. At the 
same time, the lengths of the segments as determined by LU- 
DERITZ, agree substantially with our own, when the differences 
in the total length of the cord and the great individual varia- 
tions in the lengths of the segments are both taken into ac- 
count. 

This can be shown by the accompanying Table (III), in 
which the percentage values of the lengths of the different 
regions of the cord as determined from our own Table (I), and 
from the observations of LUDERITZ on two male cords, are com- 
pared with the range in the lengths in these regions, as found 
by RAVENEL in II male cords. 


TABLE III. 


To show the percentage value of the lengths of the spinal cord within 
the regions named in the table, as compared with the range in 
these percentages as determined by RAVENEL. 


From Table (I) Liideritz Ravenel 
Region. 2 males, 2? 2 males, p. 460 II males, p. 348 
Cervical 21.7% 24.1% 19.8-25.0% 
Thoracic 55:8% 54-9% 53-2-65.4% 
Lumbar 13-9% 11.3% 9.1-13.6% 
Sacral and Coccygeal 4% 9.6% 1,8-15.2% 


For these reasons we are justified in employing our own 
data in marking off the abscissa for the curve showing areas, 
since, in so doing, we shall give an approximately true picture 
of the relations in the human cord of medium size. 

For the abscissa, it was decided therefore to take 441.6 
mm. and this base line was divided into lengths equal to the 
average lengths of the segments as recorded in Table I. 


Il, Areas of Cross-Sections. 


In the first instance charts were drawn life size, so to speak, 
with the abscissa 441.6 mm. long. The ordinates were drawn 


DoNnALDSON AND Davis, Human Spinal Cord. 25. 


on a scale of one linéar millimeter for each square millimeter in 
the area of the transverse section of the cord. The original 
charts thus made, were too large to publish, and have therefore 
been reduced photographically to exactly one-third their linear 
dimensions. As a consequence of this reduction, multiplying 
the length of the ordinates in the accompanying chart by three, 
will give a line as many millimeters long as the section has 
square millimeters of area, and multiplying the length of any 
segment in the chart by three will give the length of the seg- 
ment as represented in the column of averages in Table 1. 

The data for the areas of the cross sections represented by 
the ordinates were taken from STILLING (3) and comprise his. 
measurements on four adult cords. 


TABLE IV. 


Giving the pages in STILLING’s ‘‘Neue Untersuchungen iiber den Bau 


des Riickenmarks’”—1859, where the records are to be found. 


Designation 


of Curve. Sex. Age. Pages. 
A Male 45 years Page 1098 
B Woman 35 years Page 1100 
Cc Woman 25 years Page 1099 
D Male 25 years Page 1097 


The sections from each segment of these cords are also 
depicted in Table XXVIII of Sritttne’s Atlas (1859). 

This is the place to call attention to the condition of the 
cords measured by Stitiinc. All of the measurements of 
areas used in this present article—including those on the imma- 
ture cords to be mentioned later on—were made on material 
hardened in chromic acid and preserved in 97% alcohol. 
STILLING makes a statement of his method on pp. 1032 and 
1033, but it appears to be erroneous in that it calls for so large 
an amount of chromic acid, practically an 8% solution. Pre- 
liminary observations made on the spinal cord of the white rat, 
hardened in chromic acid 0.6%, followed by 97% alcohol, in- 
dicate a slight increase in the volume of the cord after this 
treatment. This suggests that these measurements by STILL- 


26 JoURNAL OF COMPARATIVE NEUROLOGY. 


ING may give a somewhat greater area than would appear in 
the fresh cord. 

Although the exact effect of STILLING’s treatment has yet 
to be determined, it is not to be anticipated that more than a 
small correction will need to be made for it, and as the several 
cords used were all treated by the same method, they are com- 
parable among themselves. 

In accordance with the measurements of STILLING, three 
curves were constructed for each individual. These curves rep- 
resent his determinations for the total areas of each section, as 
well as the areas of the white and the gray substance, taken 
separately ; all measured at the most caudal level of each segment. 

To obtain a general expression for these several measure- 
ments, a composite curve—the first on the chart—was made 
from the averages of the four individual records. 

This composite curve shows the maximal total area of the 
cord to occur at the VI cervical segment, the next greatest areas 
being at the III lumbar and V Lumbar, a result dependent on the 
large area of the white substance in C VI, of both gray and 
white in L III; and of the gray in L. V. 

B. ON THE VOLUME OF GRAY MATTER IN THE SEVERAL SEGMENTS 
OF THE CORD. 

A special feature of this chart as now plotted, is that it 
enables us to estimate the volume of gray substance belonging 
to the several segments. This volume was determined as fol- 
lows: 

For the first segment we multiply the number of square 
millimeters represented by the ordinate, by the number of mil- 
limeters representing the length of the segment. In the case 
of the first segment of the cord, the result is probably a trifle 
too large. Below the first, we can make a more accurate de- 
termination of the volumes by using to represent the area one- 
half of the sum of the two ordinates limiting each segment; this 
area being multiplied by the length of the segment intervening. 
Working in this way, the following results have been obtained 
for the average volumes of gray substance in the segments of 
the cord as exhibited in the composite curve on Chart I. 


DonaLpson AND Davis, Human Spinal Cord. 27 


TABLE V. 


Giving in cubic millimeters the volume of the gray substance in each 
segment of the mature human spinal cord. With the exception 
of the first cervical segment, the volume is obtained by multiply- 
ing one-half the sum of two limiting ordinates by the length of 
the intervening segment, based on the data used for the composite 
curve in Chart I. 


Volume in Volume in 
Segment cubic mm. cubic mm. 
Cervical I 129 
II 157 
III 178 
IV 220 ) 
AG 224 | Cervical segments IV-VIII 
VI 275 1220 
VII 261 [ 
VIII 240 J 
Thoracic I 177 
Il 147 
Ill 14! 
IV 148 
Vv 171 
VI 198 
VII 180 
Vill 159 
1X 156 
Xx 169 
XI 178 
XII 187 
Lumbar I 216 ) 
II 184 | Lumbar segments I-V. 
III 228 $ 1086 
1V 256 
Vv 202 | 
Sacral I 192 
II 176 
Ill 105 
IV 67 
Vv 34 
‘Coccygeal iT 12 


An examination of the foregoing Table V shows some re- 
lations worthy of remark. In the first place, the greatest vol- 
ume of gray substance is here found in the segment C VI. In 
the last five of the cervical segments, the total volume of gray 
substance is 1220 cubic millimeters, being thus decidedly 
greater than the volume of substance in the five lumbar seg- 
ments, which contain but 1086 cubic millimeters. 


28 JOURNAL OF COMPARATIVE NEUROLOGY. 


Between the two intumescentiae of the cord, the segment 
with the smallest volume of gray substance (T III) contains 141 
cubic millimeters, which is more than half that in the largest 
segment, C VI, which contains 275 cubic millimeters. 

This shows first, that there is much less difference in the 
total amount of gray substance in the successive segments of 
the spinal cord, than would appear by comparison of the areas 
of their cross sections alone (see composite curve in Chart 1); 
second, that as a matter of fact, it is the cervical enlargement 
which contains the greatest volume of gray substance although 
the area of the gray reaches its maximum in the lumbo-sacral 
region. It may not be out of place to again call attention to 
the fact that the division of the base line into equal intervals 
for the successive segments of the cord, as in the charts based 
on WOROSCHILOFF’S curves, gives a set of relations which are 
misleading ; for it necessarily suggests that the volumes of the 
segments are proportional to the areas of their cross sections. 


Cc. RELATION BETWEEN THE AREA OF THE CROSS SECTION OF 
EACH SEGMENT OF THE CORD OR THE VOLUME OF GRAY MAT- 
TER IN IT, AND THE AREA OF THE CROSS SECTION OF ALL THE 
NERVE ROOTS BELONGING TO THE SEGMENT. 


LUpERITz (6), p. 478, has published a chart based on 
STILLING’S measurements, which shows the relations between 
the combined areas of the cross sections of each of the 
four spinal nerve roots, and that of the _ cross 
section of the gray substance of the segment to 
to which they belong. The two curves representing the two 
series of areas run nearly parallel to one another. The data, 
however, on which the curves are based, are not exactly com- 
parable, for a reference to STILLING (p. 392), shows that the 
areas of the cross sections of the nerves were taken from mea- 
surements in the case of a woman of 26 years; whereas the 
areas for the cross sections of the gray matter of the cord were 
from a five-year old child. | We have drawn a second curve in 
which the areas of the gray matter for the mature cord as rep- 
resented in Table VI, are used. This curve does not fit quite 


DonaLpson AND Davis, Human Spinal Cord. 29 


so closely that for the areas of the nerves, as does the curve 
based on the areas of the gray substance in the five-year old 
cord. Especially in the thoracic and lumbar regions, the area 
of the gray substance is larger in proportion to the cross sec- 
tions of the nerves than in the case of the young cord. In gen- 
eral there appears to be in the mature cord, as contrasted with 
that of the child, a shift of the larger areas of the gray substance 
one or two segments cephalad. As a consequence, therefore, 
using data derived entirely from mature individuals, the curves 
are less similar than the figure of Liprrirz (Fig. 2, p. 478) 
would indicate. 

If we now compare the relative development of the volumes 
of the successive segments of the cord, with the areas of the 
cross sections of the spinal nerves, we find that while there is a 
fair correspondence in the cervical and in the sacral regions, 
that in the thoracic and upper lumbar regions, where the seg- 
ments have grown most in length, a great disparity exists be- 
tween the volume of the gray substance and the area of the 
cross sections of the nerves, the gray substance being much 
more abundant than we should expect. This suggests that in 
these localities, the lengthening of the segments, and conse- 
quent increase in the volume of the gray substance, is merely 
one of adaptation to the elongation of the vertebrae, and not 
accompanied by any corresponding increase in the complexity 
of its structure. 


D. GROWTH CHANGES. 


I. On the Areas of the Cross Sections of the Several Segments 
of the Spinal Cord at Different Ages. 


For this comparison we have used as a standard the rec- 
ords which were employed for the composite curve in Chart I. 
As previously explained, the data for this composite curve were 
obtained by taking the average of the observations on the four 
individuals used for the construction of curves A, B, C, D re- 
spectively of Chart I. The numerical data are given in the 
column headed ‘‘Mature”’ tn Table VI. 

To compare with the composite curve we have from 


30 JOURNAL OF COMPARATIVE NEUROLOGY. 


STILLING a series of measurements of the areas of the gray and 
white substance at the level of the several segments of three 
immature spinal cords: from a child at one year, from one at 
two years, and from one at five years. STILLING’s tables con- 
taining these records are found on pages IIOI, 1102, 339 and 
343 of his ‘‘Neue Untersuchungen tiber den Bau des Riucken- 
markes.”” 1859. 

In the following Table (VI) which contains ST1LLiNne’s mea- 
surements, it is at once seen that certain segments in the five- 
year old child were not measured. In order to make this table 
comparable with the others, an interpolation for the missing 
records, C I and C II and LI and LII, has been made on the 
assumption that in the five year old cord the areas of the un- 
measured segments would form the same fraction of the sum of 
all the areas that they do in the one and two year old cord. 
At the foot of the five year old columns in Table VI, the total 
given is the one obtained after the above interpolation. More- 
over, it will be noted that in this cord identical measurements 
are given for the thoracic segments T.III-VIII and T.IX-XI. 

Individual measurements for these several segments would, 
of course, have been preferable, but there is no reason to sus- 
pect that any serious error has resulted from the method here 
employed by STILLING. 


DonaLpson AND Davis, Human Spinal Cord. 31 


TABLE VI. 


Areas of Cross-Sections of the Human Spinal Cord at Different Ages. 
The records for the cords at 1, 2 and 5 years are copied from 
STILLING. The record for the cord at maturity gives the aver- 
ages of his four tables of measurements on adults. 


Areas of White Substance. Areas of Gray Substance. 
sq. mm. sq. mm. 
Age. Age. 
Segment. Tyr. 2yrs. 5 yrs. Mature.|I yr. 2 yrs. 5 yrs. Mature. 
Cervical. 
I B04 0307/5 esa bn O2-O2) 07-07) 7h 4-2 meme UT 7.8 
II 28.98| 30.04 5B:p4 68.57| 4.60| 4.60|1_ 12-37 ite 
Ill 22.62| 23.33 32°75|) 72-37. 4-60] 4.60 11.25) 14.14 
IV 48.08) 41.36 34.65! 74.94] 18 03] 16.26 12.73| 16.52 
Vv 45-60] 41.36 42.02| 73.97| 21.21] 16.26 19.67| 19.70 
VI 46.31} 37.12 42.02} 79.18} 20.86] 20.85 19.67| 20.32 
VII 47-37| 43-83 40.39] 71.84] 18.03] 14.14 18,24] 20.38 
VIII 39.24] 46.66 33-99| 65.30) 16.97] 12.37 13.68] 15.99 
Thoracic. 
30.75] 30,04 28.59] 63.65] 6.01] 6.36 6.97] 11.66 
II 26.51} 26.51 24.12} 53-64] 6.72} 7.78 5-32] 9.01 
III 20.85} 28.99 24.12} 52.23] 4.95| 6.36 eel) AG /he7l 
IV 21.56] 26.16 ZANE 252222) beso eAsO5 5-32} 6.89 
V 22.98] 20.50 24.12] 50.10] 5.30) 6.01 5-32} 38.74 
VI 21.56| 22.98 24.12} 45.20| 4.60] 6.36 5-32| 8.04 
VII 19.44] 24.04 24.12| 47.43) 5.66) 5.30 5-32] 6.80 
Vill 20.86) 20.50 24.12] 45.15} 6.01] 5.30 5-32] 5-92 
IX 20 15| 20.85 23-83] 40.74] 5.66) 7.78 4-56] 7.33 
X 23.68] 16.97 23-83} 43-05] 4.95] 7.07 4.56) 7.71 
XI 23.68] 23.33 23.83] 41.40] 6.36) 7.07 4.56) 8.92 
XII 22.62} 20 85 21.74] 44.18] 7.42] 7-07 6.44] 10.14 
Lumbar. 
I 22.98) 28.33); (cp cat 44-00) 7:42]. +6.54 13.51 
i 22.62| 23.68 1 ese 49.6:| 10.96} 9.90 yee rae 
III 22.98) 19.09 21.15} 48.01] 14.85) 16.61 13.26] 23.59 
IV 23.33| 22.62 22.34) 43-3c! 15.20 15.91 21.02] 24.48 
Vv 20.15| 19.80 17.07| 43.4C] 17.68] 19.80 24.89] 25.01 
Sacral. 
I 22.62} 20.86] ° 17.18) 32.34] 15.91] 21.56 23.53| 24.47 
II 20.85| 13.79 17.26] 19.44] 18.38] 22.62 23.22| 17.41 
III 10.61] 10.61 9.87| 12.25] 15.20] 13.43 17.21] 12.19 
IV 10.61} 5.66 5-97, 8.30] 9.19] 8.84 10.81| 7.86 
Vv 3.89; 4.59 2.18) 4.77) 4:60) | 3:89 6.01; 6.36 
Coccygeal. 
I 1.06] 1.77 96, 2.38) 2.83) 2.47 2.70] 3.18 
Total 766.00] 741.97} 734-80/1455.07 312.53/318.48] 334.65] 410.93 


In order to compare the records in Table VI, we use the 
sums of the total areas of the section, obtained by adding the 
sums of the areas of the gray and white substance, which are 
tabulated separately. 


32 JOURNAL OF COMPARATIVE NEUROLOGY. 
TABLE VII. 
Showing the sums of the total areas in the three immature and one 
(composite) adult cord, as tabulated in Table VI. 


Age. Sum of 31 Areas. 
I year 1078.53 sq. mm. 
2 years 1060.45 sq. mm. 
5 years 1069.45 sq. mm. 
Maturity 1866.00 sq. mm. 


These figures show that from one to five years, there is 
very little variation in the sum of the total areas. It ranges 
from 1078.53 sq.mm. to 1060.45sq.mm. Since this difference 
appears as a deficit in the two year cord, it is probably the ex- 
pression of an individual variation. The extreme cases, 1 and 
3 years, differ by less than 2%, whereas between five years and 
maturity, there is an increase in the total area of the sections 
of nearly 74%. From this it is inferred that the growtli changes 
leading to the larger area at maturity occur at some time subse- 
quent to the fifth year of life. 

The analysis can be carried a step further by comparing 
the relative areas of gray and white substance in the several 
cases. 


TABLE VIII. 


To show the percentage values of the sums of all the areas occupied by 
gray and by white substance at different ages. 


White Substance Gray Substance 
Percentage Percentage 
Sums of Areas of sums of Sums of Areas of sums of 
Age. in sq.mm. total areas in sq mm. total areas 
I year 766.00 71% 312553 29% 
2 years 741.97 70% 318.48 30% 
5 years 734-80 69% 334-65 31% 
Maturity 1455-07 78% 410.93 22% 


A glance at this table shows that during the first five 
years, the proportional value of the white substance in the sec- 
tion is about 70%, whereas at maturity it reaches 78%, the 
gray substance of course showing a correlated variation. 

It appears then that from the first to the fifth year, there 


DonaLpson AND Davis, Human Spinal Cord. 33 


is little variation in the relation between the gray and white, 
and that the change in this relation must occur at some period 
‘after the fifth year. 

In this connection, it is of interest to determine whether 
the growth changes leading to the greater total area of the 
cord segments at maturity are the result of a proportional en- 
largement in the different regions of the cord. In order to de- 
termine this, it is necessary to compute the percentage values 
of the total areas in the different regions. The results of this 
computation are shown in the following table: 


TABLE IX. 
Showing for both the mature and immature cords the percentage value 
which is represented by the total areas of the segments that 
constitute the cervical, thoracic, lumbar and sacral and coccygeal 


regions respectively. 


Region. Age. 
I year. 2 years. 5 years. Maturity. 
Cervical 39-05 36.86 36.60 37.92 
Thoracic 31.85 33-86 33-18 36.30 
Lumbar 16.52 17.02 17.42 17.69 
Sacral and Coccygeal 12.58 12.26 12.80 8.09 
100% 100% 100% 100% 


From the foregoing Table IX, it appears that after the fifth 
year, the proportional growth in area has been slightly more 
rapid in the thoracic region, and less rapid in the sacral and 
coccygeal, while in the two intumescentiae, the relations at ma- 
turity are similar to those found during the first to the fifth 
year of life. From the end of the first year then, the relative 
areas of the different regions of the cord change but slightly 
during subsequent development. The statements which have 
been made on the basis of the total area can also be repeated for 
both the gray and the white substance separately, though it is not 
deemed necessary to publish the computations, as the data for 
them are found in Table VI. 

It must be remembered, however, that from the first to the 
fifth year of year of life, the medulla spinalis is growing rapidly 


34 JOURNAL OF COMPARATIVE NEUROLOGY. 


in length. From the measurements of RAVENEL (p. 350), we 
have calculated that the length of the medulla spinalis at one 
year of age would be about 200 mm., whereas he found ina 
five-year old boy the length to be 300 mm. “Yet, despite this 
increase in length, the measurements just given show that the 
transverse diameters remain practically constant. Apparently 
we have here another example of the tendency of structures to 
grow first in their long axis before enlarging at right angles to 
it. It must be remembered, however, that we are without ob- 
servations on the changes which occur within the limits of the 
first year. Y | 

On looking at Table VI, it is to be noted that the area of 
the white substance at maturity is nearly 95 % greater than it 
was at five years, while the gray substance is only 23 % greater, 
thus showing the much more rapid enlargement of the white 
substance. This being the case, it is evident that the curves of 
WoroscHILoFF, based on a five-year old child exhibiting in 
cross section hardly more than half of the white substance in 
its cord than is present in the adult, necessarily give a false no- 
tion of the relations at maturity. 


IT. Comparison of a Curve, Representing the Areas in the Chald’s 
Cord, with the Corresponding Curve for the Cord of the Adult. 


It has just been shown that from the first to the fifth year, 
the areas of the cross sections of the spinal cord remain the 
same size. It is therefore only necessary to obtain the mea- 
surements of the lengths of the segments at some period within 
these ages in order to construct a curve for the child’s cord that 
may be compared with that for the adult. 

Liperitz (p.471) gives the length of the segments in the 
cord of a female child of three and a half years. His measure- 
ments are presented in the following Table X. 


DonaLpson AND Davis, Human Spinal Cord. 35 


TABLE X. 


Lengths of Segments of the Cord as Determined by Ltperirz in the 
Case of a Girl of Three and a Half Years. 


Segment. Length inmm. Segment. Length in mm. 
a 4.7) ( I 8.25 
cea Bey at Hee 6.25 
| III 7-OY, err ae 4.5 

Seg LW oP, fas |) NY 4.1 

an V 8.6 4 V 2.9 

El Vi 6.8 

ee veut 6.3 I Bay 
| VIII 53h el || 3-4 

By ay LULL 3.8 
pat g80 a WM 3.5 
et 7.25 to) AY 3-0 
| III 7.0 
IV 8.9 3 

aS) V 10.3 a I 3.0 

3 | VI Litg. *O 

5 1 VII 13.3 

| 

e | VIII nie 75 
ee 6 1.5 
anes 9.6 
| XI 9.3 
| XI 7.75 


As will be seen from examining the table, the measure- 
ment for the first cervical segment is lacking in the original rec- 
ord, but it has been interpolated here on the assumption that 
it would have the same proportional value as in the cord at 
seven weeks. The measurements for a cord at this latter age 
being given by LiUperirz (p. 470), it is possible to make a cal- 
culation on this basis, and the result is the number which ap- 
pears in the Table X. Upon adding the lengths of all the seg- 
' ments together, we find the length of this cord at three and a 
half years, to be 212.95 mm. For comparison with this result, 
there is available RavENEL’s table (p. 550), giving the follow- 
ing individual measurements for the length of the cord in 
children. 


'The length for the first segment is interpolated, being given the value 
of 4.7 mm, 


36 JouRNAL OF COMPARATIVE NEUROLOGY. 


TABLE XI. (From RAveENEL). 


Number. _ Total Length. 
3 Boy of 2 years 245 mm. 
4 Boy of 5 years 300 mm. 
5 Girl of 9 years 280 mm. 


From this comparison it appears that the cord here chosen 
is short, the child evidently being under size even for a female. 
However, the length 212.95 mm., is well within the probable 
limits of normal variability as judged from the variations in the 
length of the spinal cord in the adult. In using the curve, 
however, it must be remembered that in this case, the lengths 
of the segments are from a female cord that is probably short 
even for this age and sex. This is all that need be said about 
the baseline of curve E. Since the measurements for the areas 
of the sections of the cord were so similar from the first to the 
fifth year, it was thought best to choose those made on the two- 
year old child (Table VI). This record of STILLING was taken be- 
cause the measurements for all the areas are given, and because 
it is the middle one of the series of three, and hence we know 
the form and size of the cord before and after this period. 

On examining curve E in chart I, it will be noted that the 
intervals on the axis of ordinates ave egual to and have the same 
value, as in all the other curves in this chart. 

Special emphasis is to be placed on this point, and atten- 
tention is particularly directed to it, since the designating num- 
bers are smaller in size than those used for the other curves, in 
which a number is given for each twenty units only. These 
peculiarities create the illusion that the intervals on the axis of 
ordinates in curve E are smaller than in the case of the other 
curves, and for this reason, attention has been called to this 
point. 

On comparing curve E with the composite curve in chart 
I, some interesting differences appear. If the composite curve 
be taken as the standard, the following statements may be 
made. The base line, or length of cord, in E is a little less 
than half aslong. The total area (called entire section in chart) 
at no point rises above 60 sq.mm., whereas in the composite 


DonaLpson AND Davis, Human Spinal Cord. 37 


curve, it runs above 100 sq¢.mm. The intumescentiae are more 
abrupt. The enlargements of the areas in the intumescentiae 
as compared with the areas in the thoracic region, are greater. 
The maximum total areas in both the cervical and lumbar in- 
tumescentiae are further caudad, and the absolute area of the 
white substance is much less than at maturity. The most 
marked deficiency in the areas of both gray and white sub- 
stance occurs in the first three cervical segments, and especially 
in the third cervical segment. That this last feature is not an 
individual peculiarity, is indicated by the fact that the one year 
cord in Table VI (the only cord available for comparison) 
shows a similar relation. As the curve E is based on a single 
individual, no significance can be attached to minor peculiarities 
in it, but enough has been shown by the comparison just 
made, to indicate that the cord of the child differs from that of 
the adult in a number of its characters, and that the curves 
showing the areas in childhood cannot be properly used to 
show the relations obtaining at maturity. 


Conclusions. 

The foregoing observations warrant the following conclu- 
sions : 

The chart given in this article is more correct than that 
based on the curves of WorRoSsCHILOFF, since the areas of the 
gray and white substance are taken from measurements on the 
mature spinal cord, and are plotted on a base line, the divisions 
of which are proportional to the lengths of the spinal segments. 
These curves show the greatest areas at the level of C. 
VI, L. III, and L.V. The curves, however, are generalized, and 
apply to a cord of medium size, the differences due to sex be- 
ing disregarded. 

Moreover, the measurements of the areas are from cords 
which had been hardened in chromic acid, and preserved in al- 
cohol. This treatment has certainly altered the size of the 
cord, but control experiments indicate that the alteration in 
size has probably been slight. A study of chart I (see Table 
V) shows that the volume of gray substance in the intumescen- 
tia cervicalis is greater than that in the intumescentia lumbalis. 


38 JOURNAL OF CoMPARATIVE NEUROLOGY. 


There is a general correspondence between the area of a 
cross section of the gray substance at the level of any segment 
and the area of the cross sections of all the spinal nerves be- 
longing to the segment. When, however, the volume of the 
gray substance, instead of the area, is used for the comparison, 
a disproportionately large amount of gray substance is found in 
the case of the thoracic and upper lumbar segments. This is 
interpreted as indicating a passive enlargement of the gray 
substance in these segments of the cords which have been most 
elongated. 

When the cords of immature individuals are compared 
with those from adults, several important relations are brought 
to light. In the first place, the sum of the total areas of the 
cross-sections of the cords, from one to five years, is practically 
the same (see Table VII), although during the period, a con- 
siderable growth in length has occurred. During this time, 
therefore, growth in the long axis has taken place without any 
corresponding growth at right angles to the long axis. 

The form of the cord from one to five years is nearly like 
that at maturity, the difference being that in the mature cord 
the relative enlargement of the areas of the cross-sections has 
become greater in the thoracic region, but less in the sacral and 
coccygeal (see Table IX). At maturity, the relative enlarge- 
ment of the two intumescentiae is practically the same as at the 
fifth year. From the fifth year to maturity, both the 


length and the weight of the entire cord as _ well 
as the area of the cross-sections at the level of the 
several segments are increased. The sum of the areas of 


the white substance at maturity is 98% greater than at five 
years, and that of the gray substance 23% greater (see Table 
VIII). This absolute increase must represent either enlarge- 
ment of elements already completely developed, or the develop- 
ment of elements still immature at the earlier age, or some 
combination of both of these processes. Yet the failure of the 
intumescentiae to increase in their relative area in the mature 
cord (see Table IX), or in their proportional length (RAVENEL, 
p. 350), would seem to indicate that during this period there 


DoNALDSON AND Davis, Human Spinal Cord. 39 


was no increase in their relative complexity; a result, which, 
to say the least, was unexpected. 

On comparing the curves for the areas of the cord at 3% 
years (curve E), with the composite curve for the adult, in or- 
der to determine the change in the form of the cord due to 
growth, several important differences appear. In the child’s 
cord, the total areas of the sections are of course less than in 
the adult; the greater deficiency appearing in the substantia 
alba. In the child’s cord, the intumescentiae are developed 
more abruptly than at maturity, and the maximal areas appear 
further caudad; yet, despite this, the sums of the total areas 
in the cervical and lumbar portions of the cord have almost the 
same relative values (see Table IX). The most marked de- 
ficiency in the child’s cord appears in the areas of the first three 
segments of the cervical region and especially in that of the 
third cervical segment, while the entire thoracic region is less 
developed than it will be at maturity. On the other hand, the 
sacral region is arrested in its later growth and becomes rela- 
tively smaller. 


BIBLIOGRAPHY. 


(1) Krause, R und AGUERRE, J: 
Untersuchungen iiber den Bau des menschlichen Riickenmarkes mit 
besonderer Beriicksichtigung der Neuroglia. Amatomischer Anzeiger. 
Bd. 18, 1900. 


(2) WoROSCHILOFF: 
Der Verlauf der motorischen und sensiblen Bahnen durch das Lenden- 
mark des Kanninchens. SBerdchte berate Verhandlungen der Koniglich 
Sachsischen Gesellsch. der Wissenschaften zu Leipzig. Bd, 26, 1874. 


(3) STILLING, B: 
Neue Untersuchungen iiber den Bau des Riickenmarkes. Casse/, 1859. 


(4) Kapyi, H: 
Ueber die Blutgefasse des menschlichen Riickenmarks. Lemberg, 1889. 


(5) RUpDINGER, T: 
Atlas der peripherischen Nervensystems der menschlichen Kérpers 
Stuttgart, 1872. 

(6) Litperritz, C: 
Ueber das Riickenmarkssegment. Ein Beitrag zur Morphologie und 


Histologie des Riickenmarks. Archi f. Anatomie u. Entwickelungs- 
geschichte. Anat. Abtheil. Jahrgang, 1881. p. 423-495. 


40 JOURNAL OF COMPARATIVE NEUROLOGY. 


(7) RAVENEL, M: 
Die Maasverhaltnisse der WirbelsAule und des Riickenmarkes beim 
Menschen. Zeztschrift f. Anatomie u. Entwickelungsgeschichte. Bd. il. 


1877. Pp. 334-356. 


EXPLANATION OF CHART I. 


This chart represents by curves, the areas of the cross-sections of several 
human spinal cords, as well as the areas of the gray and white substance as 
they appear in each section. The base line in all the charts is just one-third 
the length of the spinal cord for which it stands, and is divided into lengths 
proportional to those of the spinal cord segments of which it is composed. For 
the adult cord, the lengths of the segments given in Table I were used in 
making the original drawings. On the ordinates one linear millimeter corre- 
sponds to one square millimeter of area. In all cases the measurement of the 
area was made up at the caudal end of the segment. In the order from above 
downwards, the curves are as follows: 

Composite Curve—Bised on A, B, C, and D, to give the average of the 
several areas in the curves named. The curves are generalized and apply toa 
cord of medium length—441,6 cm. long. The influence of sex is neglected. 
The average age of the four cases would be 33 years. 

Curve A. Man of 45 years. Data for areas from STILLING. 


Curve B. Woman of 35 years. Data for areas from STILLING. 

Curve C. Woman of 25 years. Data for areas from STILLING. 

Curve D. Man of 25 years. Data for areas from STILLING. 

Curve E. Child—data for areas from STILLING’s observations on the 


cord of the two-year old child. Length of segments from 
LUDERITZ’s observations on the cord of a three and a half 
year old girl. Cord rather short. 


Curves showing area of cross section of human spinal cord. 


Sse seo White) matter: ssnsconansensssereroneene(S rey matter. 


100 


look MMV VY VivIvinT Hot Wo vo vio va vit IX X 
A. Mati 45 yrs. 


20 


100 Bb. Woman 


x! 


» Woman 2oyrs. 


D. Man zoy rs. 


CERVICAL THORACIC 


TUIMWYVUWWIn © VW Vv WW VE KX XxX 


XI (Xi 


AUTO WY = VESTER “anv = = = = VM EX XX MM eee 


CERVICAL THORACIC LUMBAR.SACRAL 


Entire section. 


Composite curves based on A. B.C and D. 


Th WW WV TW VY! 


LUMBAR 


SACRAL 


To face page 40,] 


CHART I. 


Xi. 


NEUROLOGY. Volt. 


JOURNAL OF COMPARATIVE 


se veoume 


oA men ca mmanmtin — ——e 


iS nh Mian Soyra. 


= — 


ot vine “- i ais 7 np eee ail 
ae eee? eee i Le) so 
is 


“watvanmie ey * 8 
CERT eA, 


ier 
iP Le Pets me Pr 


tA, OH ae eee 
© 5S UN pines ail 


THE BRAIN OF THE ‘ARCHZOCETI.’ 


By Gi EcLon Smite; MSA. M.D, 


Fellow of St. John’s College, Cambridge; Professor of Anatomy, Egyptian 
Government School of Medicine, Catro. 


{A paper read before the Royal Society of London, February 12, 1903. ] 


So far as I have been able to ascertain, nothing whatever 
is known of the form of the brain or, more strictly, of the 
cranial cavity in the Archzoceti. Hence no apology is 
needed for presenting even this imperfect account of two 
cranial casts representative of this sub-order, which have come 
into my hands. 

Among the Eocene remains found in the Fayim region of 
the Egyptian desert by Mr. H. J. L. Beapneri and Dr. 
CHARLES W. ANDREwS, in Igol, there was a broken skull of 
Zeuglodon,* from which it was possible to obtain a mold, repre- 
senting the form of the greater part of the dorsal and lateral 
aspects of the brain. A plaster cast was made in the British 
Museum at the instance of Dr. ANDREWws, who kindly placed it 
at my disposal for description. 

In the following winter (1902), Mr. BEADNELL found in 
the same locality a natural cranial cast of the same size and 
general form as the artificial cast of Zeuglodon. It is obvious 
at a glance, if the two specimens be placed side by side, that 
the natural mold belongs to some member of the Archzoceti, 


1 These notes were originally intended for the Report on the Survey of the 
Fayfim, to be issued by the Egyptian Survey Department, and are now pub- 
lished separately with the permission of the Under Secretary of State for Public 
Works and Captain H. G. Lyons, Director-General of the Survey Department. 


2 C. W. ANDREWS, “‘Extinct Vertebrates from Egypt,” Part II. (Extracted 
from the ‘Geological Magazine,’ N. S., Decade IV, vol. 8, I901, p. 437,— 
Zeuglodon Osiris, Dames’.) 


42 JOURNAL OF COMPARATIVE NEUROLOGY. 


but whether to the same species or even genus as the other 
specimen must at present remain an open question. 

Mr. BEADNELL kindly placed this specimen at my disposal. 

The size and relative proportions of the different parts are 
almost identical in the two casts. Nevertheless, there are a 
considerable number of differences, some features being dis- 
played in one and not in the other, and wee versa. Many of 
these differences are obviously due to the imperfections of the 
casts, and especially to the failure of the plaster mold to repre- 
sent the true form of the brain. But there are several peculli- 
arities of the natural cast—such, for example, as the form of 
the caudal part of the cerebellum and the shape of the cerebral 
hemispheres—which are difficult to reconcile with the artificial 
mold, even if we admit that the inner face of the cranium 
(from which the latter was made) is damaged or imperfectly 
cleaned. The differences, nevertheless, are sufficiently pro- 
nounced to indicate a generic distinction between the two speci- 
mens; and in this connection it is interesting to recall a state- 
ment made by Dr. AnprReEws in his first reference to Zeuglodon, 
as ‘including apparently Damers’ Z. Osiris, and perhaps a 
second species.”’’ It would, however, be very unwise, because 
it would serve no useful purpose, and possibly lead to error, to 
found a new genus or even a new species on the evidence of 
this natural cranial cast, when our source of information con- 
cerning the known genus (Zeug/odon) is as unsatisfactory as that 
obtainable from the artificial one about to be described. More 
especially so, when it is remembered that in the case of the 
only indisputable facts (z. e., size and general form) the two 
casts are agreed. [shall therefore merely describe and attempt 
to explain the meaning of the form of the two specimens, and 
leave the question of the specific identity open for future re- 
search. 

The general appearance of the brain is extraordinarily 
peculiar (figs. 1 and 2). The shape of the anterior part of the 
natural cast (fig. 1, @ and 6) closely resembles the cerebrum of 


1 ‘Geological Magazine,’ September, 1901, p. 401. 


Exuiot Smitu, Sram of Archeocete. 43 


a Lizard greatly magnified. An anterior prismatic stalk (a), 
representing the pedunculi olfactorii, suddenly expands into a 
plump, broad, smooth mass (4), showing the form of the chief 
part of the cerebrum. The maximum breadth of the two hemi- 
spheres (fig. 1, 4) is 95 mm.; the greatest length of each (mea- 


Fic. 1.—Dorsal aspect of the natural cast described in the text. Two-ninths 
natural size. a@, olfactory peduncles; 46, cerebral hemisphere; , @, 
cerebellum ; ¢, é, fragments of skull. 


sured in front from the point where the ventral surface of the . 
olfactory peduncles appear to expand into the chief mass of 
the hemisphere) is 47 mm.; and the maximum depth is 54 
mm. Each cerebral hemisphere (exclusive of the olfactory 
peduncle) is slightly broader than it is long. 

The two olfactory peduncles are represented in the natural 
cast by a single prismatic process. This extends forward for a 
distance of 37 mm. (measured along the dorsal edge) in front 
of the point where the expansion to form the hemispheres com- 


44 JOURNAL OF COMPARATIVE NEUROLOGY. 


mences; and, as the peduncles are broken across there, it is not 
possible to estimate their total length or the shape and size of 
the olfactory bulbs. 

The coronal section formed by its anterior (broken) surface 
gives an isosceles triangle with a base measuring 8:5 mm. and 
sides of 10 mm. each. It expands as it passes backward, so 
that at its junction with the rest of the hemisphere its sides are 
each Ig mm. and its base 16 mm. in length. 


Fic. 2.—Dorsal aspect of the artificial cranial cast of Zeuglodon. Two-ninths 
natural size. a, 4,c,d,as in fig. 1. @, the dorsal rostrum, and 40’, an 
irregular boss on the cerebral hemisphere. (These are probably due to 
imperfections in the cranium.) 

In the artificial cast (fig. 2) all that represents this exten- 
sive olfactory stalk is an irregular rostrum with two small boss- 
like projections, one above the other (a and a). The cerebral 
hemispheres in the natural cast have a broad base, from which 


Exuiot SmitH, Brain of Archeocett. 45 


the sides extend upward toward the narrow dorsal surface with 
a gradual slope. In the artificial cast, however, the lateral 
parts of the hemispheres seem to be expanded into full rounded 
swellings. 

Then, again, the antero-posterior diameter of the hemi- 
sphere is much shorter (being about 13 mm. less) than it is in 
the natural cast, although the breadth of the two specimens. is 
approximately the same. It may be that the anterior parts of 
the skull, from which the artificial cast was made, are so dam- 
aged that little reliance can be placed upon the mold as an in- 
dication of the exact form of the brain. In fact, if this 
artificial cast even approximates to the form of the brain, it is 
quite certain that it did not belong to the same genus as the 
animal from which the natural cast was derived. 

In other words, as we know that the artificial cast belonged 
to Zeuglodon, the probability is that the natural cast furnishes 
the first evidence of some hitherto undescribed genus of 
Archzoceti. 

Behind the part 8, which I have just described as the 
cerebrum, there is (in the natural cast) a large, irregular mass 
of avery peculiar shape, not exactly comparable to the condi- 
tion occurring in any other brain known to me. 

Immediately behind the hemispheres (d) there is a great 
transverse bar (c) measuring 125 mm. in the transverse direc- 
tion—z.e., extending on each side 15 mm. beyond the lateral 
margin of the cerebrum (6). 

Each lateral extremity of this mass (c) is expanded to form 
a large buttress. In the natural cast these buttress-like masses 
are practically vertical, and of uniform thickness ; whereas in 
the artificial cast they are obliquely-placed, and expanded ven- 
trally. In the natural cast the mesial continuation of these 
thick lateral masses (each of which measures 30 mm. antero- 
posteriorly) becomes reduced to a bridge measuring only 5 or 
6 mm. [the exact figure cannot be stated, because a piece of 
bone (fig. 1, ¢) partially covers this region]. 

In the deep concavity behind the narrow bridge of the area 
¢ (in the natural cast) two rounded, irregular, walnut-like bosses 


46 JOURNAL OF COMPARATIVE NEUROLOGY. 


project, one on each side of the middle line (fig. 1, d@). Each 
of these is 26 mm. in diameter, and is placed so obliquely that 
its surface looks almost directly backward. Shallow but clearly 
defined furrows separate these two bodies from each other and 
from the area c. In the artificial cast there is only a very 
faintly-marked indication of these bodies (fig. 2, @). 

Ata first glance it might seem that they represent the 
whole cerebellum, in which case c would be part of the cere- 
brum! But careful examination of the natural cast renders 
such an interpretation highly improbable, and comparison with 
the artificial cast seems to finally establish the belief that the 
whole of the region marked c forms part of the cerebellum. 

It is extraordinarly difficult to accurately interpret this pe- 
culiar form of cerebellum. A comparison with other primitive 
types of cerebellum’ points to the probability that the lateral 
buttresses of the mass c represent the floccular lobes, and that 
the walnut-like mass (@) represents the cerebellar lobule which 
I have called ‘‘area C” (of. cet., ‘Catalogue,’ p. 201). 7ifieibe 
objected that the lateral buttress-like mass is much too exten- 
sive to be entirely ‘‘floccular,’’ attention may be called to the 
fact that in the large aquatic Sirenia, which have retained an 
exceedingly primitive type of brain, the floccular lobes are 
enormous in comparison with those of other mammals (of. c7z., 
‘Catalogue,’ p. 346). 

It would perhaps be difficult to find elsewhere in the mam- 
malia a greater contrast than that presented by the smooth, 
reptilian-like cerebral hemispheres of these casts and the highly 
complicated, ultra-mammalian neopallium of the recent whales, 
both Odontoceti and Mystacoceti.? And yet, if we inquire 
into the nature of the factors which have molded the form and 
determined the size of the various parts of the brain in Eocene 
times and at the present, the contrast between the brain of 
Zeuglodon and the modern Cetacea loses much of its signific- 


1 Compare, for example (‘Catalogue of the Royal College of Surgeons,’ 2nd 
edition, vol. 2, 1902), Armadillo (p. 211), Tapir (p. 311), Manatee (p. 346). 
2 Vide ‘Catalogue of the Royal College of Surgeons,’ of. ctt., pp. 348—359.- 


Evuiot SmitH, Brain of Archeocet. 47 


ance, and becomes much less peculiar, even though it may not 
be wholly explained. 

In most Eocene mammals the cerebral hemispheres were 
exceedingly diminutive in comparison with those of their mod- 
ern descendants and successors. Moreover, the bulk of the 
primitive mammalian hemisphere was composed of those parts 
(hippocampus and lobus pyriformis), which are pre-eminently 
olfactory ; in other words, the neopallium (z.e., that part of the 
pallium which is neither hippocampus nor pyriform lobe) is 
especially insignificant. It is a well-known fact that the sense 
of smell loses much of its importance in mammals of aquatic 
habits (e.g., Ornithorhynchus, the Sirenia, the Pinnipedia, and 
especially the Cetacea), and in these animals the olfactory parts 
of the brain dwindle to very small proportions. In the 
Odontoceti the olfactory bulb and its peduncle actually disap- 
pear. The Archeoceti, therefore, are subject to two factors, 
which will account in some measure for their small cerebrum. 
For, in addition to the smallness of the brain to which most 
Eocene mammals are subject, there is their aquatic mode of 
life. This causes a reduction in size of just those portions of 
the pallium which form the greater part of the Eocene hemi- 
spheres. 

In the modern Cetacea the neopallium attains to the 
greatest absolute size which it ever reaches in any mammal. 
This fact cannot, however, be considered fatal to the belief in 
the close affinity of the Archzoceti and the Cetacea, because 
the extraordinary dissimilarity between the brains in the two 
sub-orders is such as we know to have been produced by the 
operation of well-recognised causes in the long lapse of time 
which separates the dawn of the Tertiary period from the pres- 
ent day. In all mammals which lead a life ‘‘in the open’’ it 
has become a condition of their survival that the neopallium 
must increase in size in each successive generation: failing this, 
the creature must either adopt a ‘‘retired and safe mode of 
life’’ or become extinct. Numerous examples might be 
quoted in support of this hypothesis. But the case of the 
Sirenia shows us how little we really know of the factors which 


48 JoURNAL OF COMPARATIVE NEUROLOGY. 


determine the size of the brain. These creatures began the 
struggle for existence in Eocene times with relatively large 
brains, in spite of their aquatic mode of life; and they have 
been succeeded by generations ot descendants whose latest 
progeny at the present day have a brain-equipment only 
slightly superior to their earlier Tertiary ancestors (vzde Cata- 
logue, of. cit. p. 344, ef seg.). Evenif we admit that the modern 
Manatees and Dugongs lead an eminently safe and retired life, 
which is in marked contrast to the venturesome and ‘‘open’’ life 
of the whales and porpoises, much still remains to be satisfac- 
torily explained. 

Perhaps the most striking feature of the brain of Zeuglodon 
is the extreme disproportion between the size of the enormous 
cerebellum and the diminutive cerebrum. In this respect the 
fossil brain presents a most marked contrast to that of all recent 
mammals, and especially to that of the Cetacea. This relatively 
great size of the cerebellum is not peculiar to the Archzoceti, 
but is common to other extinct mammals of large size. In my 
memoir on the brain in the Edentata’ the difficulty presented 
itself of adequately explaining a similar phenomenon in Glypto- 
don; and it must be born in mind, in even attempting to do 
this, (1) that the obtrusive greatness of the cerebellum presents 
itself only in large mammals and not in lowlier vertebrates, and 
(2) that the size of the cerebellum is not proportionate to that 
of the cerebrum. In the case of Glyptodon I four years ago at- 
tempted to explain these facts in this manner. 

The development of the neopallium in mammals opens up 
the possibility of the performance of many more complex mus- 
cular acts than are possible in the Amphibia or Reptilia; these 
acts require a co-ordinating mechanism, the size of which will 
be largely determined by the bulk of the muscular masses, the 
actions of which are to be harmonised, and the extent of the 
sensory surfaces which send into the cerebellum streams of con- 
trolling impulses. A large cerebellum is being demanded by a 


1 «The Brain in the Edentata,’’ ‘Linnean Society’s Trans.,’ 2nd series, 
Zoology, vol. 7, part 7, 1899, p. 381. 


Exxiior Smitu, Lran of Archeocet.. 49 


large mammalian body, even if the cerebrum is small. I cannot 
offer any more satisfactory explanation of the magnitude of the 
cerebellum in Zeuglodon than this, 

It is clear from the foregoing that the extraordinarily great 
contrast in the appearance of the brain of the Archzoceti and 
that of the Cetacea cannot be urged as a reason against their 
kinship, when it is remembered that the operation of known 
factors is quite sufficient to explain the transformation of the 
one type into the other in the time which has separated the 
Eocene period from the present. 

Having disposed of these negative arguments, we may 
consider the positive evidence for Cetacean affinity in the brain 
of Zeuglodon. 

The shape of the cerebrum, and especially its relatively 
great breadth, is peculiar. In fact, this form of hemisphere 
rarely or never occurs among mammals, other than the Cetacea. 
I have elsewhere’ attempted to explain the shortness of the 
Cetacean hemispheres by the fact that the abortion of the basal 
(olfactory) parts of the cerebrum limits their longitudinal ex- 
tension. This, however, is not the whole explanation, because 
in many microsmatic Sirenia (Za/icore), and Pinnipedia (Osaria, 
Phoca), the hemispheres are not especially broad. The dispro- 
portionate breadth seems, in fact, to be to some extent a char- 
acteristic cf the Cetacea; and, in this respect, Zeuglodon agrees 
with them. 

The peculiar elongation of the olfactory peduncles be- 
yond the anterior extremities of the hemispheres is rarely found 
in mammals, though it is common enough in Reptiles and the 
Ichthyopsida. In fact, the exact parallel to the condition found 
in Zeuglodon occurs among recent mammals only in the Cetacea.* 
An analogous condition is found in the extinct Lemuroid 
Megaladapis [described by ForsytH Major (of. cét.)] and some 
Amblyopoda. 


1 ‘Catalogue of the College of Surgeons,’ of. czt., p. 350. 

* Full references to this are given by ForsyrH Major, ‘‘On the Brains of 
Two Sub-Fossil Malagasy Lemuroids,” ‘Roy. Soc. Proc.’ vol. 62, 1897, p. 48, 
second footnote. 


50 JouURNAL OF COMPARATIVE NEUROLOGY. 


It is not without interest to note that the two outstanding 
features of the cerebral hemispheres of the Archzoceti, even if 
their value as indices of kinship be slight, both find their near- 
est parallel in Cetacea. There are no characters of the brain 
of the modern Cetacea which can be regarded as certainly dis- 
tinctive, if we put aside such features as the extreme dwindling 
of the olfactory apparatus, and the enormous development of 
the neopallium. Both must be regarded as late acquisitions, 
not to be expected in an Eocene mammal. Under these cir- 
cumstances these slight points of positive evidence of the rela- 
tionship of the Archeoceti and Cetacea must be allowed some 
value, as reinforcing the testimony of the skeletal parts. 

If we seek to institute closer comparisons between the 
brain of Zeuglodon and of the Odontoceti and Mystacoceti with 
a view to the determination of its relationships, we are not un- 
naturally doomed to disappointment. It might, perhaps, be 
supposed by some anatomists that the absence of an olfactory 
bulb in the Odontoceti might point to a closer affinity of 
Zeuglodon to the Mystacoceti, in which a small olfactory appa- 
ratus is retained. But there is every indication that the ol- 
factory apparatus of the Odontoceti has become aborted quite 
recently. 


Fic. 3.—Ventral aspect of brain of an early feetus of Monedon. Naiural size. 
a. d., locus perforatus (area desert); 4.0., bulbus olfactorius; /.., lobus 
pyriformis. 

Thus in a specimen of the embryonic brain of the Narwhal 

(Monodon), which was given to me some years ago by Professor 

Howes, the remains of the olfactory bulb (fig. 3, 4.0.) are still 


Exviot SmituH, Braz of Archeocett. 51 


quite visible as a small umbilicate area in part of the ‘‘ desert 
of Broca (fig 3, a.d.), wherefore it follows that in the 
early embryo the olfactory bulb and peduncle develop as in all 


” 


region 


other mammals. Moreover, in all Odontoceti traces of the py- 
riform lobe are found even in the adult; and in the brain of 
Kogia greyt the rhinal fissure and the typical (macroscopically 
only) pyriform lobe are retained in a form as clearly defined as 
that of any macrosmatic mammal (fig. 4). Professor HASWELL, 
in describing this brain’ emphasises the fact that ‘‘the most re- 


Fic. 4.—Ventral aspect of left hemisphere of Cogza greyt. Reduced approxi- 
mately one-half. a.d., corpus striatum (area desert); 4.0., place occupied by 
bulbus olfactorius in feetus; /.7.@., fissura rhinalis anterior; /.7.2., fissura 
rhinalis posterior; /7.4., lobus pyriformis. 

markable feature of the [basal] region, and perhaps of the 

whole brain, is in the great depth of the ectorhinal fissure, a 

feature marking off the present form very strongly from 

Delphinus”’ (p. 438). Since his illustrations do not properly 

delineate this interesting conformation, Professor HAswELi 


1 W. A. HASWELL, ‘‘On the Brain of Grey’s Whale (ogia greyz),” ‘Lin- 
nean Society of New South Wales Proc.,’ vol. 8, 1883 (publ. 1884), pp. 437-439, 
pl. XXI. 


52 JOURNAL OF COMPARATIVE NEUROLOGY. 


kindly permitted me to examine his specimen; and Mr. J. P. 
Hirt has made me an excellent photograph (of its ventral sur- 
face), roughly reproduced the accompanying drawing (fig. 4). 
It shows the complete and quite-typical rhinal fissure and the 
characteristic pyriform lobe. In its anterior part the rhinal 
fissure is fully a centimeter deep. 

The exact reproduction of these characters of the rhinence- 
phalon in an adult anosmatic Cetacean, and the presence of the 
olfactory bulb in the foetal Narwhal, show that these toothed 
Cetaceans were certainly (and probably quite recently) derived 
from ancestors presenting the normal mammalian type of olfac- 
tory apparatus. The absence of the olfactory bulb and pedun- 
cle in the Odontoceti cannot, therefore, be considered a just 
reason for adopting the utterly improbable suggestion of a nearer 
affinity of the Archzoceti to the Mystacoceti than to the 
Odontoceti. 

Estimated by the amount of sand which it displaced, the 
bulk of the natural cast (including that of a considerable 
quantity of matrix attached to the base of the brain and some 
small fragments of bone) is 410 c.c. If the necessary correc- 
tions and estimations be made from this gross cubic capacity, 
the weight of the brain in the Archzoceti must have been con- 
siderably less than 400 grammes, and perhaps nearer 300, as 
against that of the recent Cetacea, which ranges from 455 
grammes in Kogza (HASWELL) to 4,700 grammes in Balwnoptera 
(GULDBERG). 


LITERARY NOTICES. 


Functional Changes in the Dendrites of Cortical Neurones.' 


This paper is largely a reprint of a work published in 1897 in the 
first volume of the Travaux de l'Institut Solvay, whose results the 
author now regards as fully confirmed and established. Some of these 
conclusions are as follows: 

The pyriform appendages (spines or thorns) of the dendrites of 
the cortical neurones constitute the terminal apparatus of the dendrites; 
they increase considerably the surfaces of the nerve cells and play an 
important rdle in the physiology of the brain, for in the case of severe 
disturbance the appendages disappear partially or wholly from the 
affected cells. 

Varicosities represent in the adult brain pathological modifications 
of the nerve cell. They appear abundantly only in course of grave 
disorders and are rare in the brain of the healthy animal. In the 
normal adult the dendrites of cortical neurones do not show varicosi- 
ties, but are thickly set with the pyriform appendages. 

In prolonged etherization, or electrical stimulation or fatigue of 
the cortex the pyriform appendages disappear, while varicosities are 
present, but these two phenomena are really independent of each 
other and the appendages may disappear without any trace of vari- 
cosities making its appearance. The author concludes from the disap- 
pearance of the appendages that these are motile, but is unable to 
determine the mechanism of their movement. The cortex is never 
uniformly involved in the reactions to fatigue, etherization, etc., but 
beside the affected areas are others apparently unaffected. The paper 
is followed by a list of 14 titles of papers by the same author on related 
subjects. Chyna. 


1 STEFANOWSKA, MICHELINE. Les Appendices des Terminaux Dendrites 
Cérébraux et leurs différents états physiologiques. Archives des Sciences 
Physiques et Naturelles, Quatriéme période, t. XI, May, rgor, 


ii JouRNAL OF ComPARATIVE NEUROLOGY. 


The Morphological Position of the Chorda Tympani in Reptiles. 


It is generally agreed that the Eustachean tube and tympanic 
cavity of the higher vertebrates are morphological derivatives of the 
spiracular cleft of the lower fishes. If the chorda tympani of human 
anatomy runs down cephalad of the tympanic cavity, as commonly 
taught, then the precursor of this nerve in the fishes, if such there be, 
should be pre-spiracular. ALLis } has lately called in question the pre- 
spiracular (pre-tympanic) character of the mammalian chorda and of 
course the homologous nerve cannot be sought in the fishes until this point 
is determined. Since the mammalian chorda pursues an exceedingly 
tortuous course and one difficult of interpretation, it is worth while to 
notice the conditions in the reptiles. 

VERSLUYS in his recent extensive paper ? describes and figures 
these relations in several types of Lacertilia and Rhynchocephalia and 
comes to this general conclusion: ‘‘ Die chorda tympani geht vom 
Facialis meist an der Stelle ab, wo dieser sich mit dem eben beschrie- 
benen sympathischen Nerven verbindet, das ist caudal von der Columella 
und an der inneren dorsalen Ecke des Kérpers des Quadratums. Sie 
reicht dann langs der dorsalen und vorden Baukenhéhlenwand auf der 
medialen Flache des in die Paukenhdhle vorspringenden Quadratkér- 
pers bis zum Unterkiefer.” The detailed descriptions and figures 
make it very plain that the chorda tympani of reptiles is pre-tympanic 
and therefore morphologically pre-spiracular; and in the absence of 
very definite proof to the contrary, we must assume the same condi- 
tion to prevail among the mammals also. C. Si 0aH. 


Mendel and Jacobsohn’s Jahresbericht; Fifth Issue.* 


The issue of this Annual for 1901 reaches us early in 1903 and, like 
its predecessors, is an indispensable aid to all research workers in all 
departments of neurology and psychiatry. The plan of the work is 
the same as in previous issues. 


1 ALLIs, E. P. The Lateral Sensory Canals, the Eye-muscles and the 
peripheral Distribution of certain of the Cranial Nerves of Mustelus laevis. 
Quart. Journ. Micro, Sct., XLV. 2, 1901. 


2 VerRsSLuys, J. Die mittlere und Aussere Ohrsphare der Lacertilia und 
Rhynchocephalia. Zool. Jahr., Abt. f. Amat., XII, 1899. 


3 Jahresbericht tiber die Leistungen und Fortschritte auf dem Gebiete der 
Neurologie and Psychiatrie. V. Jahrgang. Bericht iiber das Jahr 1gor. 
Berlin, S. Karger, 1902. 


ee 
_e 
—e 


Literary Notices. 


Nervous System of Myxine.' 


Dr. Holm used a variety of staining methods upon strictly fresh 
material and is therefore in a position to add many details of import- 
ance to our knowledge of this critical form. His most serviceable 
preparations are those stained by the method of Gote1 and by the 
iron-haematoxyglin of HaIpENHAIN. The histology of the entire brain 
is considered; but the results, it must be confessed, are disappointing. 
While the paper contains much of value and is carefully wrought out 
and clearly arranged, yet the author has apparently failed to make 
the most of his material. The discussion of the medulla oblongata, 
comprising about half of the paper, is especially weak, due largely, it 
would seem, to the neglect of important recent literature, especially 
that coming from England and America. oA Ab 


Taste and the Fifth Nerve.? 


The study of five consecutive cases of total removal of the Gas- 
serian ganglion by Krause’s operation shows that several weeks after 
the operation there is a total loss of taste on both the tip and the back 
of the tongue on the operated side. The author concludes that all the 
fibers of taste reach the brain by the root of the fifth nerve and that 
none of these fibers reach the brain by either the seventh or the ninth 
roots. roy Ines 


The Phylogeny of the Pallium.’® 


This volume, for which we are indebted to the kindness of Pro- 
fessor G. Eliot Smith, contains descriptions of the nervous system of 
the Invertebrata and of the brain and spinal cord of the Vertebrata of 
the collections of the Royal College of Surgeons. The Invertebrata, 
spinal cords of Vertebrata and brains of Fishes, Amphibia and Birds are 
described by Mr. R. H. Burne; the brains of the Reptilia and Mam- 
malia by Professor G. Elliot Smith, assisted in the Primates by Mr. 
W. L. H. Duckworth. 


1 HoLM, JOHN F. The Finer Anatomy of the Nervous System of Myxine 
glutinosa. Gegenbaur’s Morph. Jahrb., XXIX, 3, 1901, pp. 365-401. 


2 Gowers, W. R. Taste and the Fifth Nerve. Journ. of Physiol. XXVIII, 
4, July, 1902. 

8 Descriptive and Illustrated Catalogue of the Physiological Series of 
Comparative Anatomy contained in the Museum of the Royal College of Sur- 
geons of England. Second Edition. Vol. II. Zondow: Taylor and Francis, 
1902, pp. x, 518. 


iv JouRNAL oF CoMPARATIVE NEUROLOGY. 


The volume, we should say, is very nearly an ideally perfect cata- 
logue. With its lucid descriptions and exceptionally clear wood cuts 
it is of great value as a work of reference even to those who do not 
have access to the specimens which it describes. 

At the end of the descriptions of the brains is a summary which 
we venture to quote in full. 


The human brain is by no means the largest known to us. The 
Elephant and the Great Whales possess much larger organs, and even 
the extinct Sirenian AAytina was provided with a brain of larger abso- 
lute dimensions than that of Man. In the case of these huge animals 
the enormous mass of the brain is probably to be explained by the fact 
that the increase in size of the surface of the body necessitates a 
corresponding growth of the neopallium (to which the great proportions 
are chiefly due), which is the ultimate receptive-organ for sensory im- 
pressions. 

In the case of the human brain, however, the Anthropoid Apes 
(which approach near to Man in bodily dimensions) afford us a criterion 
as to the amount of neopallium which may be regarded as ‘‘necessary” 
(in the Family Simiidz) for the reception of impressions coming from 
such an extent of sensory surface as Man possesses. When it is re- 
membered that the largest Ape’s brain is approximately half the size of 
the smallest normal human brain, and the average Gorilla’s brain only 
about one third (approximately) the weight of the average European’s 
brain, it will then be understood how great an area of neopallium (to 
which the disproportionate size of the human and Anthropoid brains 
is chiefly due) Man possesses over and above the needs of the average 
member of the Simiidz, to serve as the physical basis (so to speak) of 
an associative memory of immeasurably greater potentialities (for stor- 
ing and comparing sensory impressions) than that of any other animal. 
The feature, therefore, which distinguishes the human from all other 
brains is the relatively enormous size of the neopallium in comparison 
with the minimum which the forces of natural selection have made a 
condition of survival in a member of the Simiidee. * 

The neopallium assumes important functions and becomes a condi- 
tion of survival for the first time in the Mammalia, and in each succes- 
sive epoch it has become incumbent upon every mammal either, on the 
one hand, to adopt some eminently safe mode of life or some special 
protective apparatus to avoid extinction, or, on the other hand, to 
“cultivate” a larger neopallium, which, as the organ of associative 
memory, would enable it to acquire the cunning and skill to evade dan- 
ger and yet adequately attend to its needs. In many of the Eocene 
Mammalia (cf. the cranial cast of Dénoceras) the neopallium is reduced 


1 IT use the term ‘‘neopallium’”’ (Journ. Anat. and Phys. vol. xxxv, 1901, p. 
431) because the other parts of the pallium, #. ¢. the hippocampus and pyriform 
lobe, do not share in thisincrease. [The significance of the term ‘‘neopallium’’ 
is explained in the article here cited. Cf. also the abstract in this JOURNAL, 
vol. XII, p. xii.—c. J. H.] 


Literary Notices. Vv 


to such diminutive proportions that the brain resembles the reptilian 
type; and in each successive generation the neopallium becomes larger 
or the creature, in self-defence, is compelled to adopt some safe form 
of life. The Aippopotamus and the Sirenia are examples of mammals 
which have not kept pace in the fierce race for neopallial supremacy 
but survive by adopting habits of life which are eminently safe. The 
condition of the human brain represents the other extreme. Here the 
neopallium has attained its maximum development, and its possessor 
has not had to seek refuge either in a retired mode of life or by any 
protective specialisations of structure either for offence or defence, but 
has attained the dominant position in the animal kingdom, whilst re- 
taining much of the generalised structural features of a primitive mam- 
mal. 

This expansion of the neopallium is general and not restricted to 
any localised areas. Thus we cannot say that the greatness of the hu- 
man neopallium is to be wholly attributed to a growth of the frontal or 
of the parietal or of the occipital areas, as various writers have main- 
tained; because all parts exhibit distinct evidences of extension. But 
some regions exhibit the effects of this general expansion more de- 
cisively than others, and many writers have assumed (quite erroneously, 
I believe) that such effects are to be attributed to localised growth. ! 
Thus there are very noteworthy evidences of growth in the region 
around the insula in the human brain, but this is probably for the most 
part an expression of the general extension in a region which lends 
itself to a clear demonstration of any increase. 

In the early mammals the olfactory areas form by far the greater 
part of the cerebral hemisphere, which is not surprising when it is re- 
called that the forebrain is in the primitive brain essentially an append- 
age, so to speak, of the smell-apparatus. When the cerebral hemi- 
sphere comes to occupy such a dominant position in the brain it is per- 
haps not unnatural to find that the sense of smell is the most influential 
and the chief source of information to the animal; or perhaps it would 
be more accurate to say that the olfactory sense, which conveys general 
information to the animal such as no other sense can bring concerning 
its prey (whether near or far, hidden or exposed), is much the most 
serviceable of all the avenues of information to the lowly mammal lead- 
ing a terrestial life and therefore becomes predominant; and its par- 
ticular domain—the forebrain—becomes the ruling portion of the 
nervous system. 

This early predominance of the sense of smell persists in most 
mammals (unless an aquatic mode of life interferes and deposes it: 
compare the Cetacea, Sirenia, and Pinnipedia for example) even though 
a large neopallium develops to receive visual, auditory, tactile, and 
other impressions pouring into the forebrain. In the Anthropoidea 
alone of non-aquatic mammals the olfactory regions undergo an absolute 


1 There is no doubt that localised hypertrophies do occur, but the funda- 
mental distinction of the human brain is the gemera/ expansion of the whole 
neopallium. 


vi JouRNAL OF COMPARATIVE NEUROLOGY. 


(and not only relative, as in the Carnivora and Ungulata) dwindling, 
which is equally shared by the human brain, in common with those of 
the other Simiide, the Cercopithecidz, and the Cebide. But all the 
parts of the rhinencephalon, which are so distinct in macrosmatic mam- 
mals, can also be recognised in the human brain. The small ellipsoidal 
olfactory bulb is moored, so to speak, on the cribriform plate of the 
ethmoid bone by the olfactory nerves so that, as the place of attach- 
ment of the olfactory peduncle to the expanding cerebral hemisphere 
becomes removed (as a result of the forward extension of the hemi- 
sphere) progressively farther and farther backward, the peduncle be- 
comes greatly stretched and elongated. And as this stretching involves 
the grey matter without lessening the number of nerve-fibers in the 
olfactory tract, the peduncle becomes practically what it is usually 
called, z. e. the olfactory ‘‘tract.” The tuberculum olfactorium becomes 
greatly reduced and at the same time flattened, so that it is not easy to 
draw a line of demarcation between it and the anterior perforated 
space. The anterior rhinal fissure, which is present in the early 
human foetus vanishes (almost, if not altogether) in the adult. Part of 
the posterior rhinal fissure is always present in the ‘‘incisura tempor- 
alis,”’ and sometimes (D. 710), especially in some of the non-European 
races, the whole of the posterior rhinal fissure is retained in that typi- 
cal form which we find in the Anthropoid Apes. When this occurs we 
can easily recognise the caudal limits of the pyriform lobe, which other- 
wise becomes confused with the neopallium. 

The hippocampal fissure is of a peculiarly consistent nature, and is 
found in all mammalian brains from Orasthorhynchus to Homo. The 
rhinal fissure is equally sus generis and almost as constant as the hippo- 
campal. <A few small mammals, such as Voforyctes, Chlamydophorus, 
Chrysochloris, and some small Chiroptera, have no rhinal fissure. 

Of the sulci perhaps the most constant is. the calcarine, which is 
found in the Marsupials (both Poly- and Diprotodont), in the larger 
Chiroptera, Galeopithecus (but not in any true Insectivore, nor, strange 
to relate, Rodent, so far as I am aware), and in all the Edentates, Car- 
nivores, Ungulates, Cetaceans, and Primates. ‘This wide distribution 
of the calcarine sulcus is not generally admitted, for most writers re- 
gard the calcar avis and the calcarine sulcus as the special prerogative 
of the Primates or even Anthropoidea, and in the celebrated contro- 
versy of 1864 the late Professor Owen strove to prove that it was con- 
fined to the human brain. It is, however, the most primitive (it may, 
however, first appear at the same time as the orbital and suprasylvian 
sulci) and widely prevalent neopallial sulcus in the Mammalia. It 
makes its appearance in most mammals (soon after the hippocampal 
and rhinal fissures have developed) as a short oblique sulcus behind 
the splenium of the corpus callosum (or in the corresponding situation 
in Marsupials), and hence it is commonly called ‘‘ splenial” (Krueg) in 
non-Primate Orders, in which its true nature has not been properly 
recognised hitherto. * 


1 Meynert and Ziehen have called the splenial sulcus ‘‘calcarine” in some 
Carnivores, without indicating any valid reasons for their views. They have 


Literary Notices. vii 


Its subsequent history varies greatly in different Orders. In the 
Carnivora, Ungulata, Chiroptera, and many other mammals the sulcus 
becomes concurrent with another element of vastly less morphological 
importance, which I have called the ‘‘intercalary sulcus.” The 
calcarine-intercalary complex forms the ‘‘splenial sulcus” of these 
brains. 

In most mammals, with the exception of the Primates, the tension 
of the growing cortex in the infracalcarine region is relieved by the 
downward extension of the calcarine sulcus to the neighbourhood of 
the rhinal fissure. 

In the Anteaters, Sloths, Pangolins, Lemurs, and Apes the calca- 
rine sulcus always remains separate from the intercalary sulcus, and the 
latter joins with the genual sulcus in the Primates to form the calloso- 
marginal sulcus. 

In many Carnivores and Ungulates (and in large mammals gener- 
ally) one or more deep sulci make their appearance behind the calcarine 
sulcus, and in most cases one of these,-which we may call the ‘‘retro- 
calcarine”’ sulcus, is deeper and more constant than the others and 
often joins the calcarine sulcus. ‘This is seen to advantage in the brain 
of the Lion, Tiger, or Seal among Carnivores, or in the Horse, Camel, 
or Ox among Ungulates. This retrocalcarine sulcus is obviously of 
very minor morphological importance in comparison with the true cal- 
carine sulcus, and this is shown by its adaptability to the varying me- 
chanical conditions prevalent in different Orders.’ In the Primates 
both the retrocalcarine (Cunningham’s ‘‘ posterior calcarine”’) and the 
true calcarine (Cunningham’s ‘‘ anterior calcarine’’) sulci tend, as a re- 
sult of the occipital extension of the hemisphere, to become horizontal 
and in most cases become concurrent. In the more rapidly expanding 
human brain it often happens that the two sulci do not exactly meet, 
as they generally do in the Apes. Cunningham is thus led to the be- 
lief that the human brain differs from the Simian brain in possessing a 
retrocalcarine sulcus; but there can be little doubt that the so-called 
*‘calcarine”’ sulcus of the Apes is really a fusion of the retrocalcarine 
and true calcarine sulci, and therefore does not materially differ from 
the human calcarine complex. If the caudal extremity of this ‘‘calca- 
rine complex” be studied in the Apes it will be found to be exceedingly 
variable and unstable, so that one cannot regard it as a part of the true 
calcarine, which is an exceedingly stable sulcus. 

By true ‘‘calcarine sulcus” I mean that depression which corre- 
sponds to or produces the calcar avis. As Flower long ago pointed 
out (Phil. Trans. 1862, p. 198, footnote), the presence of a posterior 


attempted to extend its supposed homologies to the Anthropoid pattern so far 
as to utterly discredit any value that may attach to their recognition of its 
calcarine nature. 


1 At the same time the fact that it develops in the midst of the region in 
which Vicq d’Azyr’s stripe occurs in the Primates, and which represents the 
visual ‘‘centre,” lends a special interest to this sulcus, which obviously accom- 
modates the expanding visual cortex. 


Vili JOURNAL OF COMPARATIVE NEUROLOGY. 


cornu of the lateral ventricle is not necessary for the existence of a 
calcar avis. Thus we find a free calcar in many hemispheres (those of 
Orycteropus, Thylacinus, Pteropus, for example) in which there is no 
posterior cornu. But in most mammals the calcar becomes hidden by 
a great mass of fibres (compare most Carnivores and Ungulates), and 
cannot therefore be said to exist as a projection im the ventricle; and 
yet in many large Carnivores (Phoca for instance) and Ungulates 
(Camelus) a small posterior cornu of the lateral ventricle makes its 
appearance, and with it a typical calear again becomes exposed as in 
the Primates. 

It would be strange indeed if the most constant and stable sulcus 
of the mesial surface of the hemisphere of most mammals should en- 
tirely disappear in the Primates, to be replaced by another sulcus pre- 
senting identical relations to the lateral ventricle and a similar develop- 
mental history, but without being homologous. There is an over- 
whelming mass of evidence to show that the vertical part of the sulcus 
generally called ‘‘splenial” is the direct homologue of the calcarine 
sulcus of the Primates. 

In many mammals (such as the Lion) the tension of the growing 
infracalcarine neopallium is relieved chiefly by the downward extension 
of the calcarine sulcus toward the posterior rhinal fissure, but also 
partly by certain irregular and inconstant compensatory sulci behind 
and parallel to this extension. In the Primates, however, the calcarine 
sulcus becomes very obliquely placed, not only because the occipital 
region of the hemisphere becomes caudally extended above the cere- 
bellum, but also because the elongating corpus callosum pushes back, 
as it were, the pericalcarine neopallium ; and as a result of this obliquity 
the sulcus cannot be prolonged towards the rhinal fissure, as happens 
in the Carnivora and Ungulata, so that the compensatory sulcus, which 
is known as the ‘‘collateral’”’ sulcus, attains a greatly enhanced import- 
ance, and fulfils the rdle of the ventral extension of the calcarine 
sulcus. 

A study of the variable collateral sulcus in the brain of Man and 
the Apes clearly shows its compensatory-calcarine nature. 

Another result of the occipital prolongation of the hemisphere is 
that the calcarine sulcus becomes widely separated from the intercalary 
(calloso- marginal) sulcus, to which it is joined in most mammals. The 
stages in this separation are well shown by comparing, say, the brain 
of a typical Carnivore with those of the Daubentonia and the Lemurs. 
As the result of this separation a new set of mechanical conditions pre- 
vail in the area between the calcarine and the calloso-marginal sulci; 
and, to further complicate matters, the acuate sulcus formed on the 
dorso-lateral aspect of the hemisphere by the lateral and post-lateral 
sulci (é. e. the intraparietal and its ramus occipitalis transversus re- 
spectively) becomes more and more acutely flexed as the occipital pro- 
longation occurs, so that in the Cebide the sharp-pointed apex of the 
V-shaped sulcus so-formed extends toward this region of the mesial 
wall, which is, for the reasons just mentioned, already in state of ‘‘un- 
stable equilibrium,” so to speak. As the result two sulci (which may, 


. 


Literary Notices. ix 


however, be concurrent) are formed :—(a) a ventral compensatory- 
calcarine parallel to the calcarine and the dorsal limb of the postcalca- 
rine sulcus, and (4) a vertical sulcus cutting into the dorsal edge of the 
hemisphere. The latter appears to relieve the tension of the extending 
surface in a region which is obviously influenced by the proximity of 
the ‘‘apex” of the intraparietal sulcus. In some cases (e. g. Chryso- 
thrix, D. 554) this sulcus 8 may be joined to the intraparietal, but in 
most Apes it is independent of it. The two sulci a and 6 usually over- 
lap, and in most cases the intervening gyrus becomes submerged so 
that the two elements appear to form one furrow, which is the ‘‘parieto- 
occipital sulcus.” The lattter, therefore, is a complex of two (and 
often three) new elements; it makes its appearance for the first time in 
the Anthropoidea in the region between the phylogenetically old cal- 
carine and intraparietal sulci, which are the common heritage of the 
Meta- and Eutheria. This account will explain the extreme variability 
of the parieto-occipital sulci in the human brain. 

There is yet another remnant in the Primate brain of the calcarine- 
intercalary junction (which occurs in so many mammals) in the form of 
a short sulcus above and behind the splenium, which I have called 
‘‘compensatory” (instead of Broca’s misleading title ‘‘postlimbic.’’) 

The Sylvian fissure in its complete form is found only in the human 
brain, and even in Man it often imperfect. It is really a great cleft 
upon the ventro-lateral aspect of the hemisphere formed by the meet- 
ing of the peripheral opercular lips of three sulci, which are quite dis- 
tinct in origin and in their phylogenetic history. The most stable of 
these three sulci, and therefore that which takes the chief share in the 
development of the Sylvian fissure, is that called ‘‘suprasylvian” in 
most mammals. The second is an unstable sulcus analogous to the 
pseudosylvian sulcus (that which is commonly called the ‘‘Sylvian fis- 
sure”) of the Carnivora and many other mammals. And the third 
sulcus is the fronto-orbital. 

The suprasylvian sulcus is one of the most primitive and constant 
in the Mammalian series. It is the earliest neopallial sulcus to make 
its appearance on the external surface of the hemisphere in the 
course of the development of the Carnivore, Ungulate, and 
(according to the old observations of Pouchet) Edentate 
brain, synchronising in this respect with the calcarine sulcus 
on the mesial surface. Even if its identification is not alto- 
gether sure in the Marsupialia (see the accounts of TZhylacinus, 
Macropus, and Phascolomys), we know that it is a most stable sulcus in 
the Edentata, Rodentia, Carnivora, and Ungulata. In many Mam- 
mals it is joined to the less stable postsylvian (‘‘posterior suprasylvian” 
of most writers) sulcus, which we call ‘‘parallel” in the Anthroipodea. 
In the Great Anteater, however, it usually becomes separated from the 
latter and joined to a pseudosylvian sulcus to form a Sylvian fissure, not 
unlike that found in the Lemuroidea. It is significant that in the only 
case in six hemispheres of Myrmecophaga where this junction does not 
take place, it should also happen that the suprasylvian sulcus is joined 
to the postsylvian, as in the Carnivora. In Dawubentonsa the supra- 


x JouRNAL OF COMPARATIVE NEUROLOGY. 


sylvian sulcus is always separate from the pseudosylvian, and is gener- 
ally joined to the postsylvian sulcus. In the Family Lemuride the 
suprasylvian sulcus is always (or practically always) separated from the 
parallel (postsylvian) sulcus, but numerous fragmentary sulci, and a 
backwardly-directed hook at the upper end of the suprasylvian (Syl- 
vian) sulcus, or a forwardly-directed hook to the postsylvian (parallel) 
sulcus, serve to remind us of the old link between these two sulci, 
which has been broken. 

The lower end of the suprasylvian sulcus in the Lemurs overlaps 
the upper part of a pseudosylvian sulcus (of the feline type), the gyrus 
between the two sulci becomes submerged, and the resulting sulcus we 
- now call the ‘‘Sylvian fissure.” The lower end of the suprasylvian 
sulcus can be seen in many Prosimian hemispheres emerging from 
the front of the ‘‘Sylvian complex” a short distance above the rhinal 
fissure. 

In the Apes the submerged area increases in extent and is called 
the ‘‘insula.’ It is hidden by two opercula; and a comparison of a 
large series of Ape-brains seems to clearly demonstrate that the dorsal 
limiting sulcus of the insula is no other than the suprasylvian sulcus. 

In no brain does this sulcus extend so far (in the ventral direction) 
as the rhinal fissure. In many of the larger Apes it emerges slightly 
and cuts into the anterior lip of the Sylvian fissure. In Aylobates, 
Simia, and the Anthropopithect it extends forwards upon the surface so 
as almost to reach the fronto-orbital sulcus. 

The early history of the latter sulcus is not satisfactorily known. 
It is present in an exceedingly well-developed condition in all the 
Simiidz, and in a less obtrusive form in many of the larger Apes; but, 
on the other hand, it is absent in many of the Cebidz and Cercophthe- 
cide. Such being the case, it is very surprising to sometimes find in 
the Lemurs a small sulcus, which can be no other than the fronto- 
orbital. It is impossible to say with any degree of probability whether 
this sulcus is represented beyond the limits of the Primates. The di- 
agonal sulcus of the Carnivora, Ungulata, Edentata (Bradypus, Myrme- 
cophaga) occupies a position analogous to that of the fronto-orbital in 
the Primates. 

In the Anthropoid Apes there is a pronounced tendency for the 
anterior lip of this (fronto-orbital) sulcus to become opercular and to 
extend backward over the insula, the anterior limit of which is marked 
out by the sulcus itself. 

This process of operculation may be carried very far even in 
Hylobates, Simia, and Anthropopithecus troglodytes ; and in one specimen 
of Anthropopithecus gorilla (D. 656) a very close though spurious imita- 
tion of the human condition of this region is attained. 

In the human brain this process of operculation generally leads to 
‘the complete covering of the insula. The anterior lip of the fronto- 
orbital (or, as we may now call it, anterior limiting sulcus of Reil). 
grows backward to meet the temporal operculum, and thus gives rise to 
the ‘‘stem” of the Sylvian fissure. The dorsal lip of the forward ex- 
tension of the superior limiting sulcus grows down to meet the temporal 


. 


‘Literary Notices. xi 


operculum (thus forming the anterior part of the posterior limb of the 
Sylvian fissure) and also the orbital operculum (which is the anterior 
operculated lip of the fronto-orbital sulcus). The latter meeting gives 
rise to the anterior limb of the Sylvian fissure. It often happens, 
however, that the expanding cortex in the neighborhood of the meet- 
ing place of the anterior and superior limiting sulci becomes accom- 
modated by the formation of an additional operculum—the frontal. As 
the result two anterior limbs of the Sylvian fissure (instead of one) are 
produced. 

It follows from this account that a complete Sylvian fissure exists 
only in the human brain, and that the so-called Sylvian fissure of even 
the Anthropoid Apes lacks properly-constituted anterior limbs, a small 
part of. the posterior limb, and generally also the ‘‘stem” of the com- 
plete sulcus. 

The full development of the opercula leads to the abortion of the 

upper part of the fronto-orbital sulcus in the human brain. 
._ The lateral, post-lateral, and ansate sulci of the Carnivora and 
other Mammalian Orders become in the Primates the intraparietal, 
transverse occipital, and ramus postcentralis superior respectively. It isa 
moot point whether the coronal sulcus, which is so constant and preco- 
cious in the Carnivora, Ungulata, and Edentata, forms the ramus post- 
centralis inferior of the intraparietal system. The evidence seems to 
point to the sulcus rectus and the lower part of the central sulcus as 
being the real derivatives of this furrow. 

In most Apes the region lying behind the transverse occipital sulcus 
undergoes a peculiar modification leading to the formation of a great 
operculum from the posterior lip of a new sulcus, called Simian or, as 
the Germans say, ‘‘Affenspalte.”” This does not usually occur in the 
human brain, probably because the cortical areas around the transverse 
occipital sulcus undergo a greater expansion than is the case in the 
Apes. 

I have seen, however, in the brain of an Egyptian fellah a small 
indubitable Simian sulcus like that of the Gorilla. It was separated by 
a considerable interval from the mesial plane. 

It thus happens that this region of the human brain more closely 
resembles the condition found in many of the larger Cebidz (in which . 
the opercular formation has either not begun or is only just commenc- 
ing) than that of the Anthropoid Apes. Of the latter the brain of the 
Gorilla approaches the human condition much more nearly than does 
that either of the Chimpanzee or Orang. It must be remembered, 
however, that the ‘‘occipital” and ‘‘insular” regions exhibit an extra- 
ordinary amount of variation in each of the Simiide ; the average con- 
dition of these two changing areas is much nearer the human type in 
the Gorilla than in either of the other great Apes. 

Of the other sulci of the human brain (besides those already 
discussed) the only ones which can be called ‘‘old” in the phylogenetic 
sense are the orbital and possibly the inferior frontal sulci. 

The orbital sulcus is probably one of the most primitive furrows 
in the neopallium, if not the earliest. It is the only sulcus found in the 


xii JouRNAL OF CoMPARATIVE NEUROLOGY. 


most generalised mammals, Erinaceus and FPerameles. It is a very 
constant and precocious sulcus in all the Carnivora, Ungulata, Eden- 
tata, Cetacea, and many Rodents and Galeopithecus. Most writers call 
it ‘‘presylvian” in all these non-Primate orders, but there can be little 
doubt as to its homology with the orbital sulcus, although, so far as I am 
aware, such an interpretation has never hitherto been suggested. But 
it would be strange if this (the most widespread and constant) suleus of 
the neopallium should not be represented in the Primates, and there is 
no other furrow of sufficient constancy in the pararhinal region to 
represent the presylvian sulcus of other mammals. If moreover we 
compare such brains as those of Dolichotis (Rodent), Galeopithecus (In- 
sectivore), Bradypus (Edentate), and Phascolomys (Marsuipial) with the 
Lemur’s, it is clear that the ‘‘presylvian” sulcus of the former can be 
represented in the Prosimiz only by the orbital or the fronto-orbital 
sulcus. Of these the former is not only by far the more constant of 
the two sulci, but it is also that which occupies the same position and 
relationship to the rhinal fissure as the ‘‘presylvian.’’ A comparison 
of Galago and Dolithotis shows this. 

If again we compare the behavior of the orbital sulcus in the larger 
Ungulates (e. g., the camel, horse, and ox) and Carnivores (e. g., the 
Seals), we shall find that as the hemisphere increases in magnitude (and 
more especially if at the same time it becomes more microsmatic) the 
‘‘presylvian” sulcus becomes relegated to a position alongside the anterior 
rhinal fissure exactly analogous to that occupied by the orbital sulcus 
in the Gorilla’s brain. In man the simple linear orbital sulcus becomes 
complicated by numerous side branches so as to form triradiate, 
H-shaped or other patterns; but if a large number of human brains be 
examined, the orbital sulcus will be found to consist in a very consider- 
able proportion of these cases of a single deep linear sulcus, the 
apparent branches of which are mere shallow furrows of little importance. 
Not unfrequently this sulcus joins a small anterior rhinal fissure—thus 
completing the resemblance to the junction of the ‘‘presylvian” sulcus 
with the rhinal in the Carnivora and others. 

The coronal sulcus of the non-Primate mammals may be repre- 
sented in the inferior frontal and the inferior precentral sulci of Man. 
One of the earliest sulci to make its appearance in the developing Car- 
nivore and Ungulate brain is the coronal. In the Carnivores it often 
joins the lateral sulcus, in many Ungulates it is linked to the suprasyl- 
vian, in the Pig-family it is united with the intercalary sulcus. In the 
Primates the so-called sulcus rectus exhibits a similar precocity, and 
occupies a position not unlike that of the coronal in the Ungulates and 
the primitive Viverrine Carnivores. It becomes split up in the Cebide 
and Cercopithecide into two parts, the sulcus rectus (sensu stricto) and 
the sulcus arcuatus. ‘The former develops into the inferior frontal and 
the latter into the inferior precentral sulcus. 

The problem of the exact interpretation of the central (Rolando’s) 
sulcus presents many difficulties. ‘There can be no doubt whatever as 
to the homology of the mammalian lateral with the intraparietal sulcus 
of the Primates, and the interpretation of the ansate as the 


Literary Notices. xiii 


ramus  post-centralis superior is almost as __ sure. We 
find in the carnivora and the Primates respectively a 
deep and important sulcus bearing the same relations to the ansate and 
lateral sulci. In the former we call it ‘‘crucial’” and in the latter 
‘central’; the solution thus naturally suggested is that the central sulcus 
of.the Primates represents the crucial sulcus of the Carnivora. Such a 
view has often been propounded before, and has in several instances 
been disregarded for no valid reason. Thus it has been urged (with a 
singular disregard for the facts of the case) that the crucial sulcus ‘‘be- 
longs to the mesial wall,” in spite of the patent evidence afforded by 
the Arctoid Carnivora that when the crucial sulcus becomes dissociated 
from the intercalary sulcus it often lies wholly on the dorsal surface of 
hemisphere (see the brain of the Bears, the Glutton, and in fact most 
of the Arctoidea). 

If we study the forms assumed by the crucial sulcus in the large 
Carnivores (such as the Bears and Seals) and by the central sulcus in 
the large Apes (Simiidz), we cannot fail to be struck with a striking 
parallelism, which could only be produced by the operation of similar 
factors in the two cases. Moreover, the earliest phases of the develop- 
ment of the central sulcus in the Lemurs are similar to the first rudi- 
ments of the crucial sulcus in the Viverride. 

Physiological evidence (which, however, in such matters is notor- 
iously misleading) does not altogether support such an homology. In 
the Anthropoidea the central sulcus sharply marks the exact caudal limit 
of the area of excitable cortex, whereas in the Carnivora (so the physiolo- 
gists tell us) the crucial sulcus lies in the midst of the excitable area. 

If we admit the homology of the central and crucial sulci we shall 
(by comparison with the behaviour of the latter) find an explanation of 
many features of the former. According to such an hypothesis a glance 
ata Bear’s brain will at once make intelligible the meaning of the super- 
ior genu, the caudal bend in the mesial extremity, and the tendency of 
the central sulcus in the Anthropoid Apes and Man to extend on to 
the mesial surface in front of the upturned end of the calloso-mar- 
ginal sulcus. 

In the features of its central sulcus (the relative positions of the 
genua and the behaviour of the mesial extremity of the sulcus) the 
Anthropopithect approach much nearer to Man than does the Orang or 
any other Ape.’ 

The human brain is distinguished from those of the Apes by the 
abundance of sulci between these stable and constant elements. 

The superior frontal and especially the middle frontal sulci are 
much better developed than in the Apes, and innumerable sulci develop 
in connection with these. The inferior transverse sulcus (so constant 


1 As the result of further investigations since the above was written, I 
have come to the conclusion that the crucial sulcus represents the dorsal part 
of the central sulcus and that the ventral part of the latter is formed either 
from or at the expense of (mechanically) the caudal extremity of the coronal 


sulcus. 
' 


xiv JouRNAL OF COMPARATIVE NEUROLOGY. 


in the Simiide and larger Cercopithecidz) is longer and deeper; and 
the diagonal sulcus, rarely or never seen in a well-developed form in 
the Apes, is now almost constantly present as a deep, extensive sulcus, 
lying between the anterior ascending limb of the Sylvian and the in- 
ferior precentral sulcus. 

The parietal area is notably much more variable and much richer 
in secondary sulci than it is in the Apes. 

In the temporo-occipital region the ‘‘Affenspalte” of the Apes has 
disappeared, and the depth and extent of the dorsal end of the parallel, 
the transverse occipital and lateral occipital sulci are correspondingly 
increased. ‘The inferior occipital, inferior temporal, occipito-temporal, 
and collateral sulci are usually all present in a well-developed form. In 
the Apes the deepening and lengthening of any one of these sulci in- 
volved a dwindling of its neighbour—a highly developed occipito-tem- 
poral sulcus often led to the abortion of the inferior temporal, the dis- 
appearance of the anterior end of the collateral, or the curtailment of 
the inferior occipital or vice versa ; but in the human brain there is room 
for all these unstable and mutually compensatory sulci to exist in a well- 
developed form side by side. 

The expansion of the neopallium has far-reaching effects upon 
other regions of the nervous system: the fiber systems connected with 
it become more bulky, the cerebellum becomes larger, its middle ped- 
uncle—the pons—becomes so broad that it completely covers the 
trapezoid bodies and extends down to the inferior olives. In innumer- 
able ways the whole nervous system is profoundty influenced and 
modified in structure as the result, directly or indirectly, of the attain- 
ment of the neopallium to the height of its perfection. 


Obsessions and Psychasthenia.' 


In this new work, which, like its predecessors on hysteria and on 
fixed ideas, deals with large groups of so-called neuroses, P. JANET 
brings a number of features under a definite heading, the obsessions, 
impulsions, manias, folly of doubt, tics, agitations, phobias, mysopho- 
bias, anxieties, the feelings of insufficiency, neurasthenia and the modi- 
fications of the feelings of reality. The very list of these titles gives 
us a feeling of the confusion that exists in the use of the terms, and we 
must be grateful to see a more definite entity, after the pattern of 
epilepsy and hysteria, bring a new order into these conditions 
so loosely thrown together with neurasthenia. This _ spe- 
cialized group is termed psychasthenia. The analysis of the 325 
patients has led JANET to the recognition that all these types depend on 
a lowering of the ‘‘psychological tension.” Whereas hysteria shows a 
complete suppression of ‘certain facts and an exaggeration of others, 


? Dr. PIERRE JANET, Les Obsessions et la Psychasthénie. Vol. I. Félix 
Alean, Paris, 1903. 


Literary Notices. XV 


the psychasthenia shows in the place of this narrowing down of the 
field of consciousness a lowering of consciousness in its totality, with- 
out any complete and localized gaps of anaethesia, anamnesia, paraly- 
sis, and without subconscious and subjective elements, and, therefore, 
with a feeling of the ‘‘incompleteness” or insufficiency which is usually 
absent in hysteria. 

The work is of such fundamental importance that an abstract 
would be too incomplete to be ventured upon. Suffice it to say, that 
it is a work with which all those must be completely familiar who wish 
to get out of the hazy vagueness still existing concerning the so-called 
neurotic conditions. 

The second volume will contain the clinical material on which this 
first volume is built. 

With special gratification we note that beside the cross-sections of 
these conditions, that is, the analysis of the symptom-complex at vari- 
ous times, there is also some help offered in the direction of longitud- 
inal sections, that is, the analysis of the course of the disease from a 
general clinical point of view. The relations to mental disorders are 
given with some detail, especially the occurrence of melancholia, of 
paranoic states, mental confusion and hebephrenia. 

The book is dedicated to TH. R1ispoT, who may well be proud of 
the work of his pupil. ADOLF MEYER. 


MeMurrich’s Embryology.’ 


This book is, as the title indicates, strictly a text-book of human 
embryology. It is written largely from the comparative point of view 
and is quite full on the neurological side, these chapters comprising 
about one-fifth of the volume. The work is an eminently successful 
manual, the references to comparative embryology and comparative 
anatomy tiding the reader over many difficult subjects, notably in the 
nervous system. We note a few points of criticism. 

In the subject of histogenesis advantage is taken of the foundation 
laid down by the so-called zones of His to contrast sharply the histo- 
genesis of motor and sensory roots and centers in both spinal cord and 
brain. But the author has failed to carry out this method of treatment 
as far as possible and so has missed the recognition of the important 
fact that both dorsal and ventral zones are divisible into somatic and 
splanchnic zones, whose clear understanding sheds so much light on the 


1 The Development of the Human Body. A Manual of Human Embry- 
ology. By J. PLAYFAIR McMurRICcH, Ph.D. Philadelphia, P. Blakiston’s Son 
& Co. 1902. 


xvi JOURNAL OF COMPARATIVE NEUROLOGY. 


broad morphological plan of the central nervous system. Compare the 
paper by JoHNSTON on the Functional Divisions of the Central Nervous 
System in the issue of this Journal for March, 1902. 

This failure accounts for the error on p. 408 in homologizing the 
fasciculus solitarius of the oblongata with the oval bundle and hence 
with the column of BurDAcH of the spinal cord. The components of 
the peripheral nerves are ina later section correctly given and made 
excellent use of, and a recognition of the fact that the primary and sec- 
ondary centers of these components within the brain (especially those 
of the sensory roots) are likewise clearly differentiated would have shed 
much additional light on the problems of the functional divisions or 
zones of the metencephalon. 

The differentiation of the ventral plate into somatic and visceral 
series of nuclei is clearly presented (after His) and the visceral (lateral) 
series is associated with the branchial muscles and their derivatives. 
This is the traditional view and the one which the reviewer has 
adopted in his own studies as his working hypothesis; nevertheless it 
should be recognized that it is not wholly supported by the published 
facts, especially in the most recent literature, notably the papers of 
EDGEWORTH. ‘The whole problem of the embryology of the cranial 
mesoderm of vertebrates requires a renewed investigation and thorough 
critical analysis. 

In discussing the histogenesis of the ventral spinal roots, these 
fibers are described according to the now current theories as arising 
wholly from the neuroblasts of the ventral zones of the spinal cord. In 
view of the constantly recurring evidence of the end-to-end concatena- 
tion of cells in the formation of peripheral nerves, some mention should 
have been made of the ‘‘cellular nerve” of the embryo, even though the 
hypothesis that it participates in the formation of the definitive nerve 
fiber were rejected. 

On page 433, line 14, is an obvious misprint. The word ventral 
should be inserted before the word motor. 

The excellent discussion of the components of the peripheral 
nerves I would criticise in only one point; viz. the attempt to rank the 
olfactory nerve and organ as a member of the acustico-lateralis system. 
This very doubtful homology seems to be a relic of the errors of 
BLAUE, who endeavored to show that the ‘‘olfactory buds” of some 
fishes represent organs of the lateral line system which had wandered 
into the olfactory fossa and there differentiated. The futility of this 
argument was made clear on embryological grounds by MAnprRID- 
Moreno in 1886 and later by myself in 1899 (this Journal, Vol. IX, p. 


Literary Notices. “xXvil 


396). The morphological objections are still more serious. In the 
first place, the structure of the olfactory epithelium is totally unlike 
that of the organs of the auditory and lateral line systems, where hair 
cells extending only part way through the epithelium are always pres- 
ent. The methods of stimulation of the two sets of organs are totally 
different and it is not easy to see how one could have been derived 
from the other. And finally, all of the acustico-lateralis nerves ter- 
minate in the brain in a single center with very characteristic con- 
nections, which are totally unlike those of the far distant olfactory 
center. 

At the close of the discussion of the cranial netves we find this 
excellent passage: ‘‘ From what has been said above it is clear that 
the usual arrangement of the cranial nerves in twelve pairs does not 
represent their true relationships with one another. ‘The various pairs 
are serially homologous neither with one another nor with the typical 
spinal nerves, nor can they be regarded as representing twelve cranial 
segments. Indeed, it would seem that comparatively little information 
with regard to the number of myotomic segments which have fused to- 
gether to form the head is to be derived from the cranial nerves.” 

On page 458 we read, *‘ Nothing is yet known concerning the 
development of the various forms of tactile organs,” the author having 
apparently overlooked the papers by SzyMoNnowicz in the Archiv f. 
mikr. Anatomie, 1895 and 1896. 

The adverse criticisms above are, however, relatively insignificant 
as compared with the general excellence of the discussion as a whole, 
which is clear and philosophical in design and treatment. 


QO 
sd 
x 


Motor Nerve Termini in Insects. ! 


The motor nerve terminations in the striated muscles of insects 
have been studied by various methods by many histologists, among 
whom may be mentioned RouGET, RANVIER, FOETTINGER, V. THAN- 
HOFFER, CIAcciO, BIEDERMANN, RAMON y CajaLt and R. MonrTt. 
These investigators have disagreed in many points, but the author di- 
vides them into two main classes. The first class includes those, who, 
like RANVIER, recognize in the striated muscles of insects, as in those 
of the higher animals, a Dovkre’s elevation consisting of granular pro- 
toplasmic substance containing more or less numerous nuclei. The 
nerve fiber, when it reaches this elevation, loses its sheaths, the neuri- 


1 Sulla terminazione nervosa motrice nei muscoli striati degli insetti. Pre- 
liminary note by ALBERTO AGGAZZOTTI. 


XViil JoURNAL OF COMPARATIVE NEUROLOGY. 


lemma becoming continuous with the sarcolemma, and breaks up inte 
fibrils which pass into the granular substance but do not pass beyond it 
to enter into relation with the contractile substance of the fiber. The 
second class is represented by FOETTINGER, who also recognizes a 
Doymre’s elevation under the sarcolemma, containing numerous nuclei. 
The terminal branches of the nerve fiber, however, according to Fo- 
ETTINGER, pass through the granular substance and out in different di- 
rections, finally fusing with the isotropic or intermediary disk of the 
muscle fiber. This, then, represents the theory that there is a direct 
anatomic continuity between the nerve fiber and the muscle fiber. 

The author’s investigations were made on Hydrophilus piceus and 
Melolonta vulgaris. He used the haematoxylin method suggested by Dr. 
C. Necro in 1889. The fresh muscles of the wings and legs of the living 
insects were immersed for 24 to 48 hours in DELAFIELD’s haematoxylin 
solution, washed thoroughly in water, decolorized in a weak acid mix- 
ture of glycerine, water and hydrochloric acid. They were then teased 
and mounted in a medium consisting of equal parts of glycerine and 
water. 

In the insects studied, the motor nerve undergoes dichotomous 
division of the axis cylinder, the two branches diverging 
and entering a mass of granular substance which has in profile the. 
form of a cone slightly elevated above the level of the muscle fiber. 
In the interior of the granular substance, which is stained more or less 
intensely violet, he finds three or four nuclei which appear granular 
and stain deeply, resembling the telolemma nuclei described by 
K tHNE in the motor plaques of vertebrates. He was unable however 
to find the so-called sole-nuclei also described by KUHNE. 

While the author has, by this method, confirmed the findings of 
other investigators regarding the existence of a Doykre’s elevation of 
granular substance containing nuclei, similar to those found in verte- 
brates, he has as yet been unable to establish several of the most im- 
portant points in dispute as to the structure of the motor ending in the 
striated muscles of insects. He has been unable to determine whether 
the cone-shaped eminence described by him is under the sarcolemma 
or not and he is not certain from his preparations that the nerve fibrils 
form in the granular substance a terminal arborization similar to that 
found in the motor endings in the striated muscles of vertebrates. 
These points, with the minuter structure of the Doyérk’s elevation, he 
reserves for further investigation. 

The main interest of the research seems to have centered in the 
question whether the fibrils end in the Dovkre’s elevation or pass be- 


Literary Notices. XiX 


yond it in a manner similar to that of the ultra-terminal fibers de- 
scribed by Rurrint. In the twelve preparations described and figured 
by the author, eight show one or more fine nerve fibrils separating 
themselves from the main fiber either just before it enters the granular 
substance or just after it leaves it. These secondary fibrils pass on 
for variable distances to end either on the same or on a contiguous 
muscle fiber.: He therefore concludes that the motor nerve in the 
striated muscle of insects does not as a rule endin the DovEreE’s eleva- 
tion, but in the majority of cases passes beyond this, subdividing into a 
number of fibrillae, which endin other eminences of granular matter 
either in the same or in one of the neighboring muscle fibers. 

While the author has neither, with RaNnvirrR, established the 
presence of a terminal arborization in the granular sole-plate of the 
striated muscles of insects, nor, with FOETTINGER, determined a direct 
anatomic relation between the nerve fiber and the striae of the muscle 
fiber, he seems, if I interpret his figures and descriptions correctly, to 
be inclined to favor the view of the latter that the ultra-terminal fibrils, 
at least, are often closely related to the striae of the muscle fiber. 
The structure of the motor ending seems, however, from the figures 
given, to resemble in many respects that of the terminal motor plaques 
found in the striated muscles of vertebrates and it seems probable that 
a further study of these endings by some method which more com- 
pletely stains the terminal nerve fibrils and more perfectly differentiates 
the nerve and muscle tissues will show a still greater correspondence. 

DR. LYDIA M. DEWITT. 


The Comparative Anatomy of the Brains of Lemurs and Other Mammals. ' 


Professor ELLIoT SmitH explains in his introduction that this in- 
vestigation was undertaken primarily to consider the possibility of 
homologizing the sulci of the cerebral hemisphere in different orders of 
mammals; but on account of the mass of the material to be considered 
he has found it necessary to limit this report in two ways, (1) by con- 
fining attention to two sulci, the only two which are absolutely con- 
stant in all Primates; viz., the calcarine and Sylvian; and (2) by re- 
stricting the detailed account of the sulci to the lemurs with, however, 
extensive comparisons with other mammalian orders. 

This account fills 112 pages and is illustrated by 66 text-figures 
drawn with the beautifully clear, bold outlines characteristic of the 


1 SmitH, G. ELitiot. On the Morphology of the Brain in the Mammalia, 
with Special Reference to that of the Lemurs, Recent and Extinct. TZ7vams. 
Linnean Soc. of London, 2 Ser., VIII, Part 10, 1903. 


On JOURNAL OF COMPARATIVE NEUROLOGY. 


previous contributions by this author. ‘The more important of the con- 
clusions reached are included, along with others, in the extract from 
the catalogue of the museum of the Royal College of Surgeons given 
above, and need not be summarized here, save to add one point: 
‘‘'The features of the Prosimian brain become really intelligible only 
on the supposition that the Lemurs have advanced a considerable dis- 
tance in the main stream of the evolution of the Primates and have 
then retrograded; among other manifestations of this retrogressive pro- 
cess many interesting phases of the disintegration of the cerebral sulci 
are exhibited, so that it becomes possible to recognize the constituent 
elements of many compound sulci in the Primates, and so the more 
readily to compare them with the furrows found in other mammals.” ° 
C. “5a 


Development of Lepidosiren.' 


This contribution treats of the general epidermis, buccal cavity, 
hypophysis, central nervous system and sense organs, and is illustrated 
by four excellent plates. Notable among the figures are drawings of 
the brain from different aspects at successive stages, showing also the 
roots of the cranial nerves. Typical fourth and sixth nerves were 
found and figured for the first time. 

The brain of the adult Lepidosiren closely resembles that of Pro- 
topterus. The thalamencephalon and mesencephalon do not become 
marked off from one another until relatively late and the cerebral 
hemispheres arise as two separate lateral bulgings of the wall of the 
thalamencephalon. ‘The most distinctive feature of the contribution is. 
concerned with the histogenesis of the motor nerves, of which we are 
promised a more full description later. ‘The motor nerve trunks are 
already laid down at a period when myotom and neural tube are still 
in close apposition. As development proceeds and the myotom re- 
cedes from the spinal cord the nerve trunk lengthens out, increases. in. 
thickness, and becomes ensheathed in mesenchymatous protoplasm. 
At the earliest stage observed the protoplasm of the nerve trunk is. 
continuous with that of the myotom cells, which are provided with 
tail-like processes extending into the nerve. C, Jaies 


1KERR, J. GRAHAM. The Development of Lepidosiren paradoxa. Part IIT- 
Development of the Skin and its Derivatives «, Quart. Journ. Micr. Sct., N. S., 
vol. XLVI, 1902. 


AN ENUMERATION OF THE MEDULLATED NERVE 


Ill. 


WE: 


VIII. 


PIBERS INTHE DORSAL ROOTS OF THE SPINAL 
NERVES OF MAN. 


By CHARLES INGBERT. 


With thirty-two Figures. 


(from the Neurological Laboratory of the University of Chicago.) 


Introduction. 
Historical Statement. 


Determination of the Areas of the Cross-Sections of the Dorsal Roots 
of the Spinal Nerves of Man. 


KO6OLLIKER’S Determination. 
STILLING’s Determination. 
The author’s Determination. 
Comparison of Areas. 


hw Nn & 


Determination of the Number of Nerve Fibers in the Dorsal Roots of 
the Spinal Nerves of Man. 


I, STILLING’s Estimate. 

2. Author’s Enumeration. 

3. Comparison of STILLING’s Estimate with the author’s Enumer- 
ation, 


The Number of Nerve-Fibers per Square Millimeter of the Cross- 
Section of the Dorsal Roots of the Spinal Nerves of Man. 


The Classes of the Nerve-Fibers in the Dorsal Roots of the Spinal 
Nerves of Man. : 


The Relation of the Number of Nerve-Fibers Proximal and Distal to 
the Spinal Ganglia. 


The Application of the Numerical Results to the Innervation of the 
Skin. 


Summary. 


54 JOURNAL OF COMPARATIVE NEUROLOGY. 


I. Introduction. 


This investigation was undertaken in order to ascertain 
the number of medullated nerve-fibers in the dorsal roots of the 
spinal nerves of man. It was thought that such an enumera- 
tion would assist us in determining whether we may postulate 
separate nerve fibers for each of the forms of dermal sensation, 
since it would permit us to calculate the average area of the 
skin innervated by a single nerve-fiber. 


TT. Htstorical Statement. 


Attempts have been made to determine the value of the 
roots of the spinal and cerebral nerves as pathways for nerve 
impulses both by measuring the areas of their cross-sections 
and by estimating or counting the number of nerve fibers 
in them. 

Limiting our attention to these nerves in man we find that 
the area of the cross-sections of N. opticus alone has been de- 
termined by SALzER (1880)', W. Krause (1880)’, and for the 
Nn. opticus, oculomotorius, and trochlearis by DonaLpson and 
Botton (1891). 

The determination of the number of nerve fibers was made 
by H. Rosentuat (1845)* for all the cerebral nerves except N. 
olfactorius, opticus, and acusticus, by estimations based on the 
number of fibers counted in a few squares of the ocular-mi- 
crometer. TERGAST (1872) made a determination of the num- 
ber of nerve fibers in the N. abducens, but makes no mention 
of his method. Kuunr (1879)° counted the number of nerve- 
fibers in a row representing the diameter of the N. opticus, and 
estimated the entire number by the formula 1° X 3.1416. 
Krause (1876 and 1880)* made determinations for the N. 
opticus, and (1880)" for N. oculomotorius, but does not give 
his method of estimation. 

The area of the cross-section of the roots of the spinal 
nerves has been determined by KOLLIKER (1850)" and by 
STILLING (1859)". 

The only determination on record of the number of fibers 


INGBERT, Dorsal Roots of Spinal Nerves. 55 


in the roots of the spinal nerves of man is that made by StiL1- 
ING (1859)”. 

For the purpose of comparison we shall consider the de- 
terminations of the areas of the spinal nerves in detail. 


Ill. Determination of the Areas of the Cross-Sections of the 
Dorsal Roots of the Spinal Nerves of Man. 


1. Kdolhker’s Determination. ‘The areas of the cross-sections 
of the roots of the spinal nerves, as given by KoLLIKER (1850)" 
for a man and a woman, were in all probability obtained from 
fresh tissue, as no statement to the contrary can be found. 
The roots were sectioned between the spinal cord and the 
spinal ganglia at the place where they penetrate the dura mater, 
(i. e., near the ganglia). After removing the blood vessels and 
the arachnoidea, he measured the diameters of the roots, in- 
cluding the perineurium, and from these measurements calcu- 
lated the areas of the cross-sections. That KOLLIKER was cor- 
rect in considering his results too large will be seen on compar- 
ing them with those obtained by SriLiinG and myself (Table 
I*). Although the included connective tissue is the chief 
source of error in his results, yet the fact that he calculated the 
areas from the diameters of the cross-sections of the nerve roots 
as if they were perfect cylinders is, no doubt another source of 
error worth noting. Considering a Paris line (the unit which he 
employed) equal to 2.2558 mm., I have expressed his results in 
square mm. in Table I *. 

Although he determined the areas of the cross-sections of 
the ventral as well as the dorsal roots, the latter only are neces- 
sary for our comparison. 

2. Stillng’s Determination. The material used by STILL- 
ING (1859)" in his determination of the areas of the spinal roots 
was from the spinal cord of a woman, twenty-six years of age. 
The roots had been hardened in chromic acid according to 
STILLING’s method, and sectioned by hand with a razor at the 
place where they penetrate the dura mater. The sections, 
without staining, were mounted in alcohol. By means of a 
compound microscope and a camera lucida projections of the 


56 JOURNAL OF COMPARATIVE NEUROLOGY. 


sections were made on paper and the areas determined by two 
methods: 

(a) The projections were lithographed and printed on 
paper of a uniform thickness, the weight of a square mm. of 
which had been determined beforehand by several weighings. 
The outlines of the fascicles in the projections were then cut 
out and weighed. To obtain in sq. mm. the area of the pro- 
jection of the cross-section of a given root, the weight of the 
paper so cut out was divided by the weight of one sq. mm. 
To reduce the area of projections to the true area of the cross- 
sections of the roots they were divided by the square of the 
Jinear magnification of the projections. 

(b) The outlines of the fascicles as projected on paper 
were measured by means of a planimeter and their areas found 
in sq. mm. from its readings. The areas of the projections 
were then reduced to true areas. As the results obtained by 
the two methods are practically the same, only those deter- 
mined by the planimeter are given in Table I*. His results 
for the ventral roots are also omitted as not necessary for our 
comparison. 

3. Authors Determination. ‘The material used in this in- 
vestigation was from a middle-aged man, weighing 180 lbs., 
and killed by accident. The spinal cord, together with the 
roots, was removed within eight hours after death, and hardened 
in MULver’s fluid for one month, the fluid being changed every 
day during the first part of the period of hardening. The dor- 
sal roots of the left spinal nerves were embedded in celloidin 
and sectioned at the place, between the cord and the spinal 
ganglia, where they penetrate the dura mater. 

Although we speak, of these roots as all from the left side 
of this subject, in one instance this was not the case. The 
dorsal root of Th. IX was from the right side, the one on the 
left having been injured in removal. The left dorsal root of C, 
I was wanting in this subject, andthe one used was from a dif- 
ferent man weighing about 140 lbs. This was obtained for me 
by the courtesy of Dr. PeTeR Bassor. There is, however, 


INGBERT, Dorsal Roots of Spinal Nerves. 57 


no reason to think that the results of this study are appreciably 
modified by these facts. 

The 20 thick sections were stained by WEIGERT’s hae- 
motoxylin method for the medullary sheath and mounted in 
dammar. The material was obtained by the courtesy of the 
Department of Pathology, and the sections were prepared by 
Dr. Irvine Harpesty in the year Igol. 

To determine the areas of the cross-sections of the roots, 
projections of the fascicles were made on paper by means of a 
camera lucida. The areas of these projections were measured 
by means of a ConRADI planimeter. To obtain the true areas 
of the cross-sections of the roots from the areas of their pro- 
jections the magnifying power of the microscope together with 
the camera lucida used in making them was determined by pro- 
jecting a micrometer scale ruled in millimeters upon a sheet of 
paper. In doing this the micrometer slide bearing the scale 
was placed on the stage of the microscope and so projected 
that one end of the scale was in the exact center of the pro- 
jected field (as determined by a perpendicular) and the other 
end at the periphery of the field. The projected millimeters 
were marked on the paper and their length gave the magnifica- 
tion sought. The first and second millimeters from the centre 
were each magnified 68 diameters, while the third, or the one 
near the periphery, was magnified 69 diameters. 

Primary concentric circles were made on tissue paper 
through the points representing the outer ends of the projected 
first, second, and third millimeters, and six equidistant sec- 
ondary concentric circles between the outer two of the primary 
circles. The tissue paper on which were drawn these circles 
was superimposed on the projections of the roots so that the 
centre of these circles corresponded to the exact centre of the 
projected field of vision (previously determined for the pro- 
jection of the cross-section of each root) and the fascicles found 
to lie within the first and second primary circles were consid- 
ered to be magnified 68 diameters. Those fascicles the centres 
of which fell within the Ist, 2nd, 3rd, 4th, 5th, 6th, or 7th 
space (counting towards the periphery) formed by the secondary 


58 JOURNAL OF COMPARATIVE NEUROLOGY. 


circles between the second and third primary circles were con- 
sidered magnified 68.15, 68.30, 68.45, 68.60, 68.75, 68.90, 
69.05 diameters respectively. The actual areas of the fascicles 
were then found by dividing the areas of the projections of the 
fascicles by the square of the magnification as determined 


TABLE Ila. 


Areas in sq. mm. of Cross-Sections of the Dorsal Nerve Roots of the 
Spinal Nerves of Man. 


Average 
length of Seg- KOLLIKER INGBERT STILLING 
No. of Spinal ments in mm. 
Segments according to 
DONALDSON Male Male Female 


and Davis # 


I a2 -43 203 238 
II 9.7 BES 2.33 2.18 
= III 12.4 2.81 2.14 Ieee 
= IV 14.3 1.95 2.01 1.39 
x M 12.4 3.87 2.82 2.27 
oO VI 13.8 6.04 4.65 4-41 
VII 12.9 7.06 4.72 3.76 
WANE E32 5-75 di 450s 
I 12.8 2.SI 1.79 als 
II 14.2 1.54 -97 ene 
III 78 1.00 98 0.96 
IV 20.9 1.00 .9O 73 
S V 21.9 1.08 .68 78 
a VI 23 6 1.73 .59 45 
ie VII 24.2 Ke 98 ays 
a Vill 25 17s yi! 85 
IX 23.5 1.63 .67 .87 
x 227.5 1.79 .86 82 
XI 21.4 1.84 gI .99 
XII 19.6 1.73 1-2 .88 
z I 18.3 1.90 tes Tee 
S II 12.9 DE37, 222 1.50 
g Ill 11.8 2.81 2253 2257 
is LING 10.6 3-99 2.93 3-49 
é 
\ 8.2 4-49 3.20 3-39 
I re 7-06 3-44 3.96 
= II 8.4 3-99 1.92 5.52 
5 III Wat I.U1 1.18 2EG8 
Ss IV 6.7 48 .40 1.24 
V 4.8 23 a0 0.47 
2) 
5 I ov 04 03 II 


INGBERT, Dorsal Roots of Spinal Nerves. 59 


above. The measurements of each fascicle were made inside 
of the connective tissue sheath. The total area for the cross- 
section of each root is found by adding together the areas of 
its component fascicles, and is recorded in Table I*, while the 
area of each fascicle of the different roots is found in Tables 
TI-XXXII. 

4. Comparison of Areas. On comparing the areas of the 
cross-sections of the dorsal roots as determined by KOLLIKER, 
StiLLinG, and the author, and recorded in Table I? we find them 
as follows: 


KOLLIKER (male) 79.74 mm. 
STILLING (female) 57.95 mm. 
Author’s case (male) 54.93 mm. 


K6LLIKER’S areas, it is seen, are considerably larger than those 
obtained by Srituine and those by the author. This, no doubt, 
is due chiefly to the amount of connective tissue included by him 
in determining the areas from the diameters. In looking over 
my projections in Figs. 2-32 it will be evident at a glance that 
the amount of connective tissue between the fascicles is a factor 
of great importance in such a determination. 

To test the correctness of this conclusion I have deter- 
mined by KoLLIkErR’s method the area of the cross-sections of 
the dorsal roots used in my investigation. In doing this I took 
as their diameters the mean of the two diameters of the projec- 
tions. The square of the half of this diameter, i. e., the square 
of the radius multiplied by 3.1416, gives the areas of the pro- 
jections and this, in turn, divided by the square of the magnifi- 
cation of the projections, or by 68’, gives 79.14 mm’ as the 
true area of the cross-sections of all the left dorsal roots. This 
harmonizes well with KOLLIKER’s result, which was 79.74 mm? 
for the'male. The source of the difference in our ‘results ‘as 
given in Table I* is therefore evident. 

On comparing STILLING’s areas with those obtained by the 
author we again find a difference, but not so marked as that be- 
tween KO6OLLIKER’s areas and my own. Although STILiine’s 
areas are for a female and mine for a male, his are the larger. 
We again find the amount of connective tissue included to be a 


60 JOURNAL OF COMPARATIVE NEUROLOGY. 


probable source of error. On examining the projections of the 
cross-sections of the dorsal roots in Sritiine’s Atlas (1859)”, 
Table XVIII, we find that he has divided the roots of the 31 
left spinal nerves into only 228 fascicles, while, as will be seen 
from my plates, I divided the same number of roots into 504 
fascicles. It is therefore evident that he has included in his 
results the areas of some connective tissue septa which I have 
been able to exclude. Another source of difference is the fact 
that I made corrections for fascicles cut obliquely. In doing 
this, the number of fibers in a fascicle cut obliquely was di- 
vided by the number of fibers per sq. mm. in the nearest ad- 
joining properly cut fascicle providing this had the same ap- 
pearance as regards the size and density of its fibers. By this 
means the area of the fascicle in question was reduced to the 
size of its cross-section when cut at right angles to the fibers 
composing it. In computing the total area of the nerve this 
corrected number was the one employed. Where corrections 
were made the number indicating the uncorrected area is placed 
in parenthesis in Tables II-XXXII. In the accompanying 
projections, Figs. 2-32, however, all the fascicles appear drawn 
with the area which they present in the section. But where it 
it has been necessary to make a correction in the Tables for the 
obliquity of the fibers the letter C appears within the fascicle. 

The total amount thus deducted from the apparent area of 
all the roots was 1.389 mm.” Adding this to 54.93 we get 
56.39 mm.” as the uncorrected area. If to this we could add 
the area represented by the connective tissue septa which I 
have excluded but which Sritiine included, my results would 
probably be a trifle larger than his. 

Another source of difference in the results obtained by 
K6LLIKER, STILLING, and the author is the amount of shrinkage 
caused by the reagents employed. The roots used by KoLtt- 
KER were, as before remarked, most probably.not treated with 
any reagents and should in this respect be normal. STILLING 
attributes a slight shrinkage to the tissue which he measured, 
but if this be greater or less than than the shrinkage from the 
reagents I used there is at present no means for deciding 


INGBERT, Dorsal Roots of Spinal Nerves. 61 


although it seems probable from the investigations of DoNnaLp- 
SON (1894) on changes caused in the nervous tissues by re- 
agents, that the departure from the normal is but slight. 

In order better to compare the areas of the cross-sections 
already discussed, I have constructed curves. In Fig. 1, 
Chart I, (page 72) I have constructed curves to compare the 
absolute areas of the cross-sections of the dorsal roots determined 
by KoLLiker for a male, by Sti_i1nG for a female, and by the 
author fora male. In constructing these curves the normal average 
length of the spinal cord, 441.6' mm.,was used as the base line. 
On this line was marked the normal average length of each 
segment, as determined by Donatpson and Davis (1903).™ 
The ordinates were erected at the caudal end of each segment 
and each sq. mm. cf the area of the cross-section of each dorsal 
root was represented by 20 mm. on the axis of the ordinates. 
For publication this chart has been reduced so as to make its 
base line 96 mm. which is a reduction to 0.22 of the original 
linear dimensions. In Table I* are found the data for this 
chart. 

In Chart II, (page 72) I have compared the same data as are 
used in Chart I. The base line is the same as before. The ordi- 
nates, however, were in this case obtained by taking the largest 
area in each instance as 100% and expressing the other areas 3 
as percentages of this standard. Each percent. in the values 
thus obtained is represented by 1 mm. on the axis of the or- 
dinates. For publication this chart has been reduced as in 
Chart I. In Table I? are found the data for this chart. 


1 The discrepancy between 441.6 here used and 441.9 in table 14 is due to 
the fact that the latter is the sum of the numbers after being rounded from two 
decimal places to one. 


62 JoURNAL OF COMPARATIVE NEUROLOGY. 
TABLE Ib. 
Showing Percentage Values of the Areas of the Cross-Sections of the 


Dorsal Nerve Roots as Given in Table I@. The Greatest Area in 


each Series is taken as 100%, and the other Areas calculated 
from this as a Standard. 


Percentage Value of Areas. 


No. of Spinal KOLLIKER INGBERT STILLING 
Segments Male Male Female 
if 6.0 2.5 5-9 
II 53-1 45.6 27-2 
= Ill 39.8 41.8 27.5 
= IV 270 39.8 252 
‘ V 54.8 55et 4l.1 
oS) Wali 85-5 90.9 79.9 
VII 100.0 92.3 68.1 
VIII 81.4 100.0 72.6 
1 39-8 35-2 45-5 
II 21.8 18.4 24.7 
Ill 14.1 19.1 17.2 
IV 14.1 17.6 1352 
i) V 15.3 13 3 14.1 
4 VI 24.5 11.5 13.6 
i) VII 24.5 19.1 13.2 
S VIII 24.5 13.7 15.7 
IX 23.0 leit 15.4 
x 25.3 16.8 14.8 
XI 26.0 17.8 1720 
XII 24.5 24.4 15-9 
ci I 30.9 34.2 22.1 
Ss I 33-5 43-4 27-1 
g Il 39.8 49-5 46.5 
5 IV 56.5 7.3 63.2 
V 63.6 62.6 61.4 
I 100.0 67.3 71.7 
‘S II 56.5 YS 100.0 
9 Ill 16.4 23.0 45-5 
n IV 6.7 7.8 22.4 
V Bee 25 8.5 
Co 
3S I 5 6 19) 
IV. Determination of the Number of Nerve Fibers im the 


Dorsal Roots of the Spinal Nerves of Man. 


The further significance of the areas of the cross-sections 
of the dorsal roots can only be fully appreciated when the num- 
ber of fibers which these roots contain has been determined. 
As before mentioned, there is, up to the present time, on record 


InGBERT, Dorsal Roots of Spinal Nerves. 63: 


only one attémpt at a determination of this kind, viz., that of 
STILLING. 

zr. Stiling’s Estimate. The material used by STILLING 
(1859)” for this work was the same as that employed in mea- 
suring the areas. In this investigation he made use of a com- 
pound microscope in the oculus of which was placed an ocular 
micrometer on which was a square 3’” (Paris lines) on each 
side, and each side again was divided into 120 parts by parallel 
rulings. | He counted in a certain number of the small squares 
thus formed the number of cross-sectioned nerve fibers seen 
within them, and from this estimated the number for the entire 
square. This he found to be from 120-176, average 148, for 
the dorsal roots. He does not state what roots he examined 
nor how many counts were made. Since the magnification 
used was 37.5 diametérs or 1406 in area, the amount of the 
section seen within the square would be g-1406 or I-156 of a 
square line. If 1-156 of a square line of the section contains 
148 fibers, one square line will contain 156X148, or 23,088 
fibers, which is equal to 4,537 fibers for each sq. mm. Since 
in his case there were 57.95 mm.” in the area of the dorsal 
roots of the 31 left spinal nerves, there will be 57.95 X4,537, or 
262,919 nerve fibers in all of them. And since the right dorsal 
roots have an area of 55.09 ram.”, they will have 55.09 4,537, 
or 249,943 nerve fibers, and the dorsal roots of both sides 
5 12, 862. 

STILLING’s estimate for the number of nerve fibers in the 
dorsal roots of both sides is, however, not 512,862, but 504,- 
473. This difference is due to the fact that he based his esti- 
mate on the area of the dorsal roots as obtained by his method 
of weighing, which gave 111.23 mm.” for both sides, and the 
calculations I have made are based on his areas as obtained by 
the planimeter, which gives 113.04 mm.” The latter areas 
were selected because they may be more properly compared 
with my own, which were also obtained by planimetric mea- 
surements. 

2. Author's Enumeration. In making the enumeration of 
the number of nerve fibers in the dorsal roots of the left spinal 


64 JOURNAL OF COMPARATIVE NEUROLOGY. 


nerves of man, I made use of the same material as was em- 
ployed in determining the areas of the cross-sections of these 
roots. The fascicles in the projections used for determining the 
areas, as is shown in Figs. 2-32, were numbered. Photo- 
graphs of some 200 diameters enlargement were made from 
small negatives of the cross-sections of the roots, and in these 
the fascicles were numbered as in the projections; in addition 
the sub-divisions of the fascicles were lettered. These sub- 
divisions were not made into individual fascicles, because the 
septa between them were not sufficiently well developed to war- 
rant it. Both the projections and the photographs were placed 
on the table in front of the microscope and used in cen 
the fascicles the fibers of which I wished to count. 

The sections were well stained and the fibers easily recog- 
nized in all buta few instances. The characteristic mark sought 
for in case of doubt is the ring-like appearance of a nerve fiber 
due to the transparency of its axis-cylinder. In rare cases, 
however, where a nerve fiber was sectioned through its node of 
RANVIER this characteristic was not prominent. By means of 
this characteristic a small nerve fiber is easily distinguished 
from a connective-tissue corpuscle or a fragment of the medul- 
lary sheath. Some difficulty was found in the _ obliquely 
sectioned fascicles because the axis-cylinder in such cases was 
not easily recognized. The amount of connective tissue in 
some of the thoracic roots was also a source of difficulty in that 
it made the sections denser by filling up the spaces between the 
nerve fibers, and thus tended to obscure the small fibers. It is 
not thought, however, that any important errors have been 
made as the result of the difficulties named. 

In counting, a ZEISS microscope fitted with a mechanical 
stage, a ZEISS objective, 4 mm., aperture 0.95, and oculus No. 
6 or 8, was used. Oculus No. 6 was used only in a few in- 
stances where a coarse-fibered fascicle or one of its sub-divisions 
was too large for the field of vision ina No. 8 oculus, but not 
too large for that in an oculus No. 6. 

In the oculus was placed an ocular micrometer ruled into 
square millimeters. The cross-sectioned nerve fibers seen with- 


INGBERT, Dorsal Roots of Spinal Nerves. 65 


in the upper left-hand square were always counted first, then, 
in the same row, the squares to the right of this, and each suc- 
ceeding row below the first in alternate directions, so as to 
avoid confusion by passing across the fascicle from the last 
square in one row to the first in the next row. The counting 
was done by means of an automatic register, having a counting 
limit of 999, and worked by the thumb of the right hand. 
When a sub-division had been counted the number indicated 
by the register was recorded on a paper which was numbered 
and lettered to correspond to the photograph of the cross- 
section of the root counted. The sub-divisions were later 
added and the totals placed after the number of the fascicles, as 
is shown in Tables II-XX XII. 

When one sub-division was too large to be seen within one 
field of the ocular net, the method employed (except in a few 
cases where oculus 6 was used as mentioned above) was to find 
landmarks in the section and similar ones in the photograph. 
One of the limiting lines of the ocular micrometer was made to 
lie across these landmarks, which in most instances were two 
conspicuous nerve fibers; one on each side of the fascicle. This 
limiting line was so arranged as to be at right angles to one of 
the movements of the mechanical stage. After counting the 
fibers within all the squares on the side of the limiting line 
within the field of vision, the mechanical stage, carrying the 
slide with it, was moved by means of its screws, and the oppo- 
site limiting line made to run through the same landmarks, and 
the rest of the sub-division counted. In counting the cross- 
sectioned nerve fibers crossed by one of the lines of the ocular 
micrometer only those fibers the axis-cylinders of which were 
entirely within that side of the line first counted were consid- 
ered as belonging to that side, while the rest were considered as 
belonging to the other side of the line and counted 
when that side was counted. Most of the counting was done 
by daylight from a north window. Some counting, however, 
was done by the aid of an incandescent light placed two feet 
in front of the microscope. 

After counting for fifteen or thirty minutes a few minutes 


66 JoURNAL OF COMPARATIVE NEUROLOGY. 


were taken for rest, and after one hour of counting a longer 
rest was taken. On the average, about two hours a day were 
devoted to counting, though on several occasions the period 
was longer. 

The number of fibers counted per hour varied greatly. 
On three or four mornings, under the most favorable condi- 
tions, as many as 3,000 fibers were counted per hour. Under 
ordinary conditions, from 1,000 to 1,500 fibers per hour was 
the best that could be done. In very many instances less than 
500 per hour was found to be a severe task. 

In estimating the accuracy of this enumeration, three 
sources of error have to be considered. 

(a) Errors in the identification of the small nerve fibers. 
This has already been mentioned. The sources of error here 
are that either connective tissue corpuscles, or that fragments 
of broken medullary sheaths may be mistaken for nerve fibers. 
The characteristic sought for in doubtful cases of this kind is 
the ring-like appearance of the nerve fiber due to the greater 
transparency of its axis-cylinder. The only case where this 
test is not conclusive is where the fiber is cut so obliquely as to 
fail to show this ring-like aspect. But in these cases the 
elongated appearance of the fiber is quite different from the ap- 
pearance of a connective tissue corpuscle and markedly differ- 
ent from the sharp-angled fragments of broken medullary 
sheaths. Since in only a comparatively few instances are the 
fibers either cut obliquely or broken into fragments, I consider 
this source of error very small. 

(b) Errors in counting, due to the overlooking of some 
fibers, or to counting the same fibers twice. To determine the 
amount of this error I counted, preparatory to my enumera- 
tion, certain fascicles several times and found the results to vary 
less than 2%. It must be apparent that this error also in- 
cludes, at least to a large extent, the error from source (a). 
Still it is probable that the error from this source is a trifle 
greater than this preparatory test shows, because of greater 
fatigue at times, and poorer conditions, such as light, at other 


IncBERT, Dorsal Roots of Spinal Nerves. 67 


times. Ido not believe, however, that the error from sources 
(a) and (b) exceed 2% on the average. 

(c) Errors due to the possibility that the smallest fibers 
were not stained, and consequently not seen. The fact that 
my count gives aresult 60% greater than STILLING’s result 
seems to me good evidence that the smallest fibers were in- 
cluded. Again, by measurements, I found. fibers the diameter 


TABLE Te. 


Showing the Number of Nerve-Fibers Counted in the Dorsal Roots of 
the Left Spinal Nerves of Man. 


Number of Nerve Number of Nerve 


No. of Spinal Area of Roots Fibers in Each Fibers in thousands 
Segments in sq. mm. Root per sq. mm. 

I 13 1808 13.6 
II 2.33 28375 12.2 
S gin! 2A 27119 12.6 
ps IV 2.01 27102 13.4 
fe Vv 2.82 28204 10.0 
oe) VI 4.65 46549 10.0 
Vil 4.72 50278 10.6 
VIII Salat 50173 9.8 
I 1.79 17891 10.0 
II 97 13432 13.7 
Ill .98 11701 11.9 
IV -90 11375 12.5 
5 Vv .68 8352 ne 
s ae 59 7155 TIE 
2 -95 12325 12.5 
ae VIII sii 8983 12.6 
IX 67 8163 12.1 
x .S6 10612 12.3 
XI -gI 11403 12.4 
XII £.25 14125 11.4 
a I 1,75 18861 10.8 
= II 2522 23640 10.6 
g III 2.53 31328 12.3 
5 LY, 2.93 39653 13.5 
Vv 3.20 43128 13.4 
I 3-44 47461 13.8 
3 II 1.92 25545 ies 
g Ill 1.18 17322 14.9 
n IV .40 $580 21.3 
Vv ai 2223 19.3 

8 
5 I 03 761 16.0 


Totals 31 54.93 653,027 11.9 


68 JOURNAL OF COMPARATIVE NEUROLOGY. 


of which was only about 24% to be quite numerous. I think, 
therefore, that the smallest medullated fibers are included in 
my count. 

In Tables II-X XXII are to be found the number of nerve 
fibers for each fascicle, and in Table I° is found the totals for 
each root. 

3. Comparison of Stilling’s Estimate with the author's Enu- 
meration. On comparing the determination of the number of 
nerve fibers in the dorsal roots of the spinal nerves made by 
STILLING with those of the author a wide difference is manifest. 
According to his estimate there are only 262,919 fibers in 
these roots on one side, while according to my count there are 
653,627. In other words, STILLING’s result is 39.9%, or in 
round numbers 40%, of my results. In searching for the 
source of this difference I found two significant statements by 
STILLING (1859). First, that he finds no nerve fibers of so 
small a diameter as 2.7—4.5, and second, that the nerve fibers 
range, in diameter, from 7—22y. It thus seems evident that 
for the most part STILLING failed to see nerve fibers the diame- 
ters of which were much less than 7y, and in this lies the chief 
source of the difference between our results. I find, as we 
have seen, that STILLING’s results are about 40% of mine, or in 
other words, my total, in round numbers, is 60% greater than 
the results of Stititinc. It therefore remains to be determined 
whether this 60% of difference can be explained by the fibers 
the diameter of which is less than 74—these being the fibers 
smaller than STILL1NG’s lowest limit. 

In searching for data to determine this point I find that 
RosENTHAL (1845)‘, BrnpER and VoLKMaANN (1842), KOLLIKER 
(1850), and DucHENE (1864)", estimate the small fibers in the 
dorsal roots of the spinal nerves of various vertebrates to range 
from 4%—¥% of their entire number. 

SIEMERLING (1886-1887)”-, counted in each dorsal root, 
mostly on the left side, the nerve fibers seen in 9 mm.’ of the 
ocular micrometer. He found that 670 of the nerve fibers so 
counted were less than 5 in diameter, and 536 between 5. 3— 
23.94—that is, 55% % are small and 44% % large fibers. In 


INGBERT, Dorsal Roots of Spinal Nerves. 69 


order to determine, roughly, this relation between the large and 
small fibers | have counted all the nerve fibers the diameter of 
which is more than 7y in the dorsal roots of C.II and Th. VII 
of the spinal nerves of the left side. According to this count 
the dorsal root of C.II contains 12,621, and that of Th. VII 
4,621 of these large fibers. From Table I¢ it is seen that the 
former contains in all 28,375, and the latter 12,325 nerve fibers. 
Calculations based on these data indicate that in the dorsal root 
of C. II 55.5% of the fibers are ess than 7y, and in the dorsal 
root of Th. VII, 62.4%. 

We therefore conclude from the calculations based on the 
results of StTiLLinG and on my own results, that 55 to 60% of 
the fibers in the dorsal roots of the spinal nerves are less than 
7 in diameter, i. e., having a smaller diameter than the small- 
est nerve fiber measured by Stitiinc._ The failure on the part 
of STILLING to include the fibers the diameters of which are 
less than 7 is no doubt the chief source of the difference be- 
tween STILLING’s results and my own. 


V. The Number of Nerve Fibers per Square Millimeter of the 
Cross-Section of the Dorsal Roots of the Spinal Nerves of 
Man. 


According to the data which we have taken from STILLING 
the dorsal roots of the spinal nerves in man have an area on the 
left side of 57.95 mm.” and contain 262,gI9 nerve fibers, or in 
other words, they contain 4,537 nerve fibers per sq. mm. of 
the cross-section. 

According to my results, the dorsal roots in the case ex- 
amined have on the left side an area of 54.93 mm.’, and con- 
tain 653,385 nerve fibers, or on the average, I1,gOI nerve 
fibers per sq. mm. of the cross-sections of the roots. 

This comparison is but another way of showing that SriLt- 
ING failed to observe the small fibers. 

In my projections of the dorsal roots of the left side, as 
well as in the tables accompanying them, are found the relative 
number of nerve fibers for each fascicle, expressed in thousands 
persq. mm. From a study of these it is evident that the rela- 


70 JOURNAL OF COMPARATIVE NEUROLOGY. 


tive number may vary considerably even in the same root. It 
follows, in general, that the fascicle having a relatively large 
number of fibers per sq. mm., other things being equal, con- 
tains the fibers of smaller diameter. This holds true in a great 
number of instances, as can easily be seen from a microscopical 
examination of these fascicles. Still other factors beside the 
caliber of the nerve fiber enter in to determine this relative 
number. One of these factors is the amount of connective tis- 
sue present in the fascicles. This varies greatly in the different 
roots due, perhaps, to the fact that the roots were not sectioned 
at exactly the same point with regard to the dura mater. Thus, 
the cross-sections of the thoracic roots show far more connect- 
ive tissue than the cervical and lumbar roots. This, no doubt, 
is one reason why the lumbar roots, although they have a 
great number of large fibers, yet show such a large number of 
fibers per sq. mm. On the other hand, this is most probably 
also the reason why the thoracic roots, which contain many 
small fibers yet do not show a higher relative number per 
sq. mm. 

Another factor is the compactness of the tissue of the 
fascicles. That this is often determined by the connective tis- 
sue is evident; still in many cases the fibers may be seen to be 
much scattered without the presence of much connective tissue ° 
between them. 

Upon studying my results in order to determine whether 
there be any relation between the diameter of the fascicle and 
its number of nerve fibers per sq. mm., although no conclusive 
statement can be made respecting this point, yet it can be said, 
that in general the small fascicles have fibers of a small caliber. 

On the. basis of the results obtained by the measurement 
of the roots and the enumeration of its fibers I have con- 
structed Chart III in Fig. 1. This has been constructed in the 
same manner as Chart II. The percentage values for these 
curves are found in Table I“ and the figures on which these 
values are based in Table I°. 


InGBERT, Dorsal Roots of Spinal Nerves. 71 


TABLE Id. 


Giving the Percentage Values of the Areas of the Cross-Sections of the 
Dorsal Roots of the Spinal Nerves as Compared with the Per- 
centage Distribution of the Number of Medullated Nerve Fibers 
in Each Dorsal Root. The Greatest Value in each Series is 


taken as 100%. 


Percentage Value. (Ingbert) Male 
a a Nn ce eee 


Number of Number of 

Segments Areas Fibers 
ee eR ee 

I 3.6 2.5 

II 56.4 45-6 

3S ill 53-9 41.8 

a IV 53-9 39-3 

os Vv 56.0 55-1 

Oo VI 92.5 90.9 

VII 100.0 92.3 

VIII 99.7 100.0 

I 35-6 .2 

Il NOn7/ 3 

III 2252 19.1 

IV 22.6 17.6 

= V 16.6 1323 

S VI 14.2 11.5 

‘ig VII 24.5 19 1 

= VIII 17.6 13,7 

ID,< 16.2 nih 

D4 21.1 16.8 

XI 22.6 17.8 

XII 28.0 24.4 

6 I 37:5 34.2 

S II 47.0 43-4 

5 Ill 62.3 49.5 

3 IV 78.8 Hine: 

V 85.7 62.6 

I 94.3 67. 

: um 50.8 375 

g III 34.4 23.0 

n IV 17.0 7.8 

V 4-4 2.5 

12) 
8 I 1.0 6 


92 JOURNAL OF COMPARATIVE NEUROLOGY. 


Chart I 
A.—— Male. Kolliker. 
Be as von Wi qle 6 Ing bert. 
C.———— Female. Stilling. 


= ae a ete ES 
Slee ae ene, 


Kolliker. 
Inéb ert. 


de e. \ 


\ 


ITTV VY UWWInM VY v Wo vt Wl K X M XT I 0 mwvinmre 
CERVICAL THORACIC LUMBAR SACRAL 


Fig. 1 


Chart I. Curves showing the areas in sq. mm. of the cross-sections of the 
dorsal roots of the left spinal nerves. Each sq. mm. is represented by 20 divi- 
sions on the axis of ordinates. 


Chart JJ. Curves based on the measurements used in Chart I; the values 
in each curve being entered in percentages of the greatest area, which is taken 
as 100%. One mm. on the axis of ordinates equals 1%. 


Chart /I1. Curves showing in percent. the areas of the cross-sections of 
the dorsal roots of the left spinal nerves (Author), and the percentage distribu- 
tion of the number of medullated nerve fibers in each dorsal root. (Author). 
The close concordance between these two curves is a matter of interest as well 
as of importance, as it shows that there is a close relation between the size of a 
root and the number of nerve fibers in it. The divergence in the Lumbar 
region I take to be due to the compactness of these roots at the places where 
they were sectioned. 

The depression in the curve above C. IV is also a fact worth noticing, 
whatever may be its significance. 


INGBERT, Dorsal Roots of Spinal Nerves, 73 


VI. The Classes of Nerve Fibers in the Dorsal Roots of the 
Spinal Nerves of Man. 


That the dorsal roots of mammals are pathways for affer- 
ent nerve impulses has been considered as demonstrated since 
the time of Bett (1811)”. This fact has been so often con- 
firmed in man as the result of injury and disease as to need no 
further discussion. But whether the dorsal roots of man con- 
tain efferent nerve fibers in addition to the afferent, is yet an 
open question. The histological studies of v. LENHOSSEK 
(1890)*, Caja (1890) and (1893)”, VAN GEHUCHTEN (1893)”, 
ReEtzius (1892)”, and Martin (1895)”, have demonstrated in 
the chick the existence of neurones, the cell bodies of which 
are located in the spinal cord and the axones of which pass out 
of the cord by the dorsal roots. It is inferred that these neu- 
rones are efferent, but the demonstration of this is still lacking. 

By physiological experiments STEINACH (1893 and 1898)”, 
STEINACH and WIENER (1895) and Horton Situ (1897)" 
have demonstrated the presence of efferent fibers in the dorsal 
roots of the frog. 

The degeneration method has given contradictory results. 
According to VEjas (1883)” and JosEPH (1887)*, this method 
demonstrates in the cat and the rabbit the presence of fibers 
which degenerate towards the ganglion after section of the dor- 
sal root, while by the same method Kanter (1884)”, SINGER 
and Minzer (1890), SHERRINGTON (1894)” and (1897)*, VAN 
GEHUCHTEN (1895), and Gasri (1896)” found no evidence of 
such fibers in the cat, dog, or monkey. 

STRICKER (1876)", Bonuzzi (1885 and 1887)”, GARTNER 
(1889), Morar (1892)*, Hasreriik and Bigeprt (1893)*, 
WERZILOFF (1896)", and BayLiss (1900 and 1902)** have 
demonstrated by physiological methods vaso-dilator fibers in 
the dorsal roots of the spinal nerves of the dog. Roux (1900) 
in man in cases of tabes reports the demonstration of degener- 
ated nerve fibers of a small caliber in the ram communicantes. 
In order to determine the location of the trophic centres of 
these fibers Roux sectioned in the cat the dorsal roots of sev- 


74 JOURNAL OF COMPARATIVE NEUROLOGY. 


eral thoracic nerves and found this to cause a degeneration of 
small fibers in the sympathetic trunk. He therefore concludes 
that the cell bodies of these neurones are located in the spinal 
cord. It is thus evident that as regards man we have not yet 
any satisfactory demonstration of efferent nerve fibers in the 
dorsal roots, although their existence seems possible. 


VII. The Relation of the Number of Nerve Fibers Proximal 
and Distal to the Spinal Gangha. 


On this problem we have yet no data for man. SHER- 
RINGTON (1894-5) and others make the statement that in mam- 
mals a medullated nerve fiber in the periphery branches little or 
not at all except near its termination. Harpesty (1899 and 
1900)’, and others have shown that in the frog the number of 
nerve fibers distal to the spinal ganglia is greater than the 
number in the dorsal and ventral roots combined. Ac- 
cording to Dunn (1g00)”, in the frog about 6 to 8% of 
all the fibers in the mixed nerves which innervate the 
thigh divide; and according to her results on the frog, pub- 
ished in 1902”, in the mixed nerves; about 9-10% of the 
fibers to the thigh and about 21-22% of the fibers in the shank 
divide. That nerve fibers may divide at places where the 
nerve trunk gives off no branches is also demonstrated by her 
work, 

For our calculations, however, it is immaterial whether or 
not an afferent fiber branches, because the relation which we 
shall try to establish is one between the area of the dermal 
surface and the nerve fibers at a place where they are /east nu- 
merous, i. e., in the dorsal roots. Even if there should be a 
distal excess due to some other cause than distal branching, 
and even if the fibers which form such an excess should be 
pathways for afferent impulses, yet the dorsal roots would still 
constitute the only pathway by means of which these afferent 
impulses could reach the spinal cord. 


INGBERT, Dorsal Roots of Spinal Nerves. a5 


VILL. The Application of the Numerical Results to the Innerva- 
tion of the Skin. 


The author had hoped at this time to publish his estimate 
of the innervation of the dermal surface of the human body 
based on the enumeration here made. In making such an esti- 
mate the number of afferent fibers innervating the skin and the 
muscles respectively was calculated by the aid of results ob- 
tained by Sri_tinGc and VorscuviLto. But since this part of 
the thesis was written the author has commenced an enumera- 
tion of the number of nerve fibers in the ventral roots of the 
spinal nerves of man. This investigation as far as carried out 
shows that the relation of the number of nerve fibers in the 
dorsal, and the ventral roots as given by STILuine is not suffi- 
ciently accurate for an estimate of the innervation of the skin. 
The publication of this part of this investigation will therefore 
be deferred until the number of nerve fibers in the ventral roots 
of the spinal nerves of man has been determined, thus present- 
ing proper data for calculating the percent of the dorsal root 
fibers which innervate the skin and the muscles respectively. 


LX. Summary. 


1. The total area of the cross-sections of the dorsal roots 
of the left spinal nerves of a large man is 54.93 mm.” 

2. The total number of medullated nerve fibers in the 
dorsal roots of the left spinal nerves of the same man is 65 3,- 
627; and the total number on both sides would therefore be 
about 1,307,254. The error in this instance is probably less 
than 2%. ' 

3. There are, on the average, 11,900 medullated nerve 
fibers to every sq. mm. of the cross-sections of the dorsal roots 
of man. 

4. There is a close relation between the area of the cross- 
sections of the dorsal roots and the number of nerve fibers 
which they contain. (See Chart III, Fig. 1.) 

5. The small fascicles of a dorsal spinal root, in general, 
contain nerve fibers of small caliber. 


76 JouURNAL OF COMPARATIVE NEUROLOGY. 


6. The number of nerve fibers per sq. mm. of the cross- 
sections may vary considerably in the different fascicles of the 
same dorsal spinal root. 


In the following Figures (2-32) are given the projections of the 
dorsal roots of the spinal nerves of man. Their magnification is 34 
diameters. Within the outlines of each fascicle are found two num- 
bers; the first designating the number of the fascicle, and the second 
the number, in thousands, of the nerve fibers per sq. mm. of its 
cross-section. Thus in C.I there is but one fascicle and this has 13.6 
thousands or 13,600 nerve fibers per sq. mm. of its cross-section. In 
a few of the outlines of the fascicles is found the letter ‘‘c.” This is 
to signify that the fascicle was found to be cut obliquely, and that the 
number indicating the nerves per sq. mm. is that of some neighboring 
fascicle, and that the corrected area of this fascicle was obtained by 
dividing the number of fibers in the fascicle by this number of fibers 
per sq. mm. The outlines are always those of the uncorrected pro- 
jections. In the Tables (II-X XXII) accompanying the projections 
are found the data for each root. These tables have four columns of 
figures; in the first, is found the number given to each fascicle; in the 
second, the area in sq. mm. of the cross-section of each fascicle; in 
the third, the number of nerve fibers counted in the cross-section of 
each fascicle; and in the fourth, the number, in thousands, of nerve 
fibers per sq.mm. of the cross-section of each fascicle. In cases 
where a fascicle was found to have been cut obliquely, and its area 
has been corrected, the number indicating the corrected area is placed 
in the column itself (the second) and the number indicating the uncor- 
fected area after it in parenthesis. At the bottom of the table are found 
the totals. These totals have been brought together in Table I° . 


ae 


ue y : © 


TABLES AND FIGURES. 


OURNAL OF COMPARATIVE NEUROLOGY. 


INGBERT, Dorsal Roots of Spinal Nerves. 79 


SPAS IE By oD Coe ire 


Area of Fascicle 


Number of Fascicle in sq. mm. 


I 


0.13 


Number of Nerve 


Number of Nerve 


Fibers in each Fibers in Thousands 


Fascicle 


1.808 


ANC ANN (CN 


Areaof Fascicle 


per sq. mm. 


13.6 


Number of Nerve Number of Nerve 
Fibers in Each Fibers in Thousands 


Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I .0528 558 10.5 
2 -0549 728 13.2 
3 -1076 1241 11.5 
4 -0692 707 10.2 
5 0153 222 mea 
6 -0397 479 12.0 
7 -0529 548 10.3 
8 .O715 749 10.5 
9 .0716 744 10.4 
10 .0743 913 12-3 
11 10755 941 12.4 
12 -0555 807 14.5 
13 .0870 1299 14.9 

14 .O157 319 20.3 
15 .1410 1589 Ge) 
16 . 1047 1315 12. 
17 .0926 1271 13.8 
18 -1349 1876 13.9 
19 .1926 2221 11.5 
20 -1641 2055 12.5 
21 -1202 1596 12.5 
22 .0884 1036 Meh 7 
23 .0366 557 15.2 
24 .1636 1854 13-3 
25 -2452 2750 Kea 
‘Totals, 25 2.3274 28,375 12.2 average 


OURNAL OF COMPARATIVE NEUROLOGY. 


INGBERT, Dorsal Roots of Spinal Nerves. 81 


EAB Vs Ce tele 


Number of Nerve Number of Nerve 
Areaof Fascicle Fibers in Each Fibersin Thousands 


Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I -0268 276 10.2 
2 .0302 592 19.6 
3 .0478 673 14.1 
4 -0786 1018 12.9 
5 .0782 gIo 11.6 
6 0254 345 13.5 
7 -0303 460 11.9 
8 -O112 197 17.6 
9 .O155 262 16.9 

10 .0229 397 16.4 
II .0438 615 14.1 
12 .0257 508 19.7 
13 .0848 855 10.0 
14 -0252 356 14.0 
15 -0990 1005 10.1 
16 .0612 980 16.0 
17 .0308 438 14.2 
18 .O415 534 12.8 
19 -O118 183 ESS 
20 .0521 592 i123 
21 .0644 826 12.8 
22 -0940 1054 U2 
23 -0574 549 9-5 
24 -O128 215 16.8 
25 .O189 309 16.3 
26 -0267 416 15.5 
27 .0556 601 10.8 
28 .0446 652 14.6 
29 -0530 488 Q.2 
30 - 1066 1551 14.5 
31 -0220 396 19.5 
32 .0777 1021 Ian 
33 -1236 1288 10.4 
34 .0518 570 Il.o 
35 -1008 1215 E21 
36 .0498 743 14.9 
37 .0781 837 10.7 
38 . 1800 2077 DELS 
39 0355 386 10.9 
40 -0286 335 9 yf 
41 -0249 394 15.8 


Totals 41 2,1396 27,119 12.6 average 


82 JOURNAL OF COMPARATIVE NEUROLOGY. 


INGBERT, Dorsal Roots of Spinal Nerves. 83 


ABLE Vien Cp lvi. 


Number of Nerve Number of Nerve 
Area of Fascicle Fibers in Each Fibersin Thousands 


Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I .1090 1335 E2:2 
2 -0526 654 12.4 
3 10555 691- 12.4 
4 -1580 1627 10.3 
5 .2869 3158 11.0 
6 -1810 2166 II.g 
7 -0994 1237 12.4 
8 .2011 2868 14.2 
9 .0187 282 I5.1 

10 .0909 1458 16.0 
TI -0532 937 17.6 
12 .0304 685 22.5 
13 -0528 997 17.0 
14 -0273 439 16.1 
15 .0153 204 1353 
16 1043 1625 15.5 
17 .0598 897 15.0 
18 .0972 1518 15.6 
19 -1615 2225 1a 7 
20 - 1081 1560 14.4 
21 -0078 79 10.2 
22 .0398 460 11.5 


Totals 22 2.0106 27,102 13.4 average 


Journat or Comparative Nevrotoey. 


INGBERT, Dorsal Roots of Spinal Nerves. 85 


ADN 3 bd MESS (Geic 


Number of Nerve Number of Nerve 
Area of Fascicle Fibers in Each Fibersin Thousands 


Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I .0194 214 11.0 
2 .0608 721 11.8 
3 .0837 1311 15.6 
4 .0837 825 9.8 
5 -1144 (.1217) 1017 8.9 
6 -0821 1050 12.7 
7 .0613 803 13.1 
8 -0956 1099 tals 
9 -1364 1169 8.5 
10 .1388 1349 9.7 
Ii .O821 611 7.4 
12 .0754 822 10.9 
13 -0316 321 Io.I 
14 -0871 872 10.0 
15 0550 499 9.7 
16 -1055 gol 8.5 
17 :0259 345 13.7 
18 .0357 410 11.5 
19 .03381 466 11.9 
20 .0998 (.1008) 1006 10.0 
21 -1112 1036 9.2 
22 .2270 (.2310) 2014 8.8 
23 1425 1218 8.5 
24 .1241 1142 9.2 
25 .0602 (.0802) 594 9.8 
26 .1692 (.1816) 1445 8.5 
27 .1265 (.1596) 1075 8.5 
28 -1072 DEL2 8.5 
29 .0972 (.1274) 1196 12.3 
30 .10SI 1229 123) 
3! -0319 332 10.1 


Totals 31 2.8180 28,204 10.0 average 


INGBERT, Dorsal Roots of Spinal Nerves. 87 


TABLE VII. C.VI. 


Number of Nerve Number of Nerve 
Area of Fascicle Fibers in Each Fibers in Thousands 


Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I -I415 1583 I1.1 
2 -1532 1647 10.7 
R .0782 898 nie 
4 .1956 1724 8.8 
5 2717 2327 8.5 
6 . 1409 1198 8.5 
i .1526 1299 8.5 
8 .0937 942 10.0 
g -1957 Tor 9-7 

10 . 1468 1683 11.4 
II .1379 1310 9.5 
12 .1675 1365 7.1 
13 -3462 3233 9.3 
14 .O917 967 10.5 
15 -1279 1862 9.5 
16 .1046 1181 Whee 
17 .0237 280 11.8 
18 .1013 1032 10.2 
19 .0890 905 Dist 
20 .0956 1007 10.5 
21 .0944 1103 11.6 
22 .1406 1312 9.3 
23 .1709 1491 8.7 
24 1791 1770 Dos) 
25 .1460 1454 9.9 
26 .1052 1058 10,0 
27 OZ 1391 7.9 
28 .2899 2854 9.8 
29 02h2 1361 le 2 
30 .0840 1165 13.8 
31 .1136 1536 13.5 
32 .1539 1700 11.0 


Totals 32 4.6533 40:549 10.0 average 


Journat or Comparative NEUROLOGY. 


INGBERT, Dorsal Roots of Spinal Nerves. 89 


TABLE, Vil Cviil: 


Number of Nerve Number of Nerve 


Area of Fascicle Fibers in Each Fibers in Thousands 
Number of Fascicle in sq. mm. Fascicle per sq. mm. 

I -0626 639 10.2 

2 .1303 1329 10.2 
B .0738 858 11.6 
4 2119 2403 Hi 1fa3} 

5 -2045 2656 12.9 
6 2808 2788 95 

Gh -I1410 1632 11.5 
8 S206 2052 9.7 
9 1432 1442 10.0 
10 .1852 1706 ¢ 9.2 
II -1197 1536 12.9 
12 1039 1397 13-4 
13 .O751 920 12 2 
14 0579 949 16.2 
15 .OQI2 1219 13.3 
16 2111 1962 9.3 
17 3411 2598 7.4 
18 .2877 2435 8.4 
19 .0O81 571 8.4 
20 .2077 1684 8.1 
21 -1597 1276 7-9 
22 .0806 1423 20.1 
23 -2477 2439 9.8 
24 .0433 628 14.2 
25 .0778 638 $8.2 
26 .1163 953 8.2 
27 .1805 1871 10.3 
28 .0532 740 11 8 
29 .1393 1609 139 
30 -1O14 1852 11.5 
31 .0864 1256 18.1 
32 .1318 1673 14.5 
33 .0970 1144 1257, 


Totals 33 4,7233 50,278 10.6 average 


go JoURNAL OF COMPARATIVE NEUROLOGY. 


INGBERT, Dorsal Roots of Spinal Nerves. gI 


TABEEY EX) Covi. 


Number of Nerve Number of Nerve 


Area of Fascicle Fibers in Each Fibers in Thousands 
Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I -2573 2231 8.8 
2 .1736 1586 9.1 
3 3133 2993 9:5 
4 -1176 i322 11.2 
5 .0680 601 8.9 
6 -0226 331 14.0 
7 sii gis 1274 Ligig72 
8 0569 802 14.0 
9 .O501 791 15.8 
IO E22 2323 10.0 
II .0733 673 9.1 
12 .0612 515 8.4 
13 .0721 714 9.9 
14 .1797 1495 8.3 
15 .1582 (.1710) 1301 8.6 
10 .1527 (.2140) 1314 8.6 
17 .0838 722 8.6 
18 .0241 360 14.8 
19 -2091 1897 9:7 
20 0452 409 9.0 
21 - 7007 $395 11.9 
22 .1659 1694 10.2 
23 .1695 (.2096) 1637 10.2 
24 .1399 (.1718) 1366 10.2 
25 .1642 (.1675) 1490 10.2 
26 2511 2237 8.9 
27 .1961 1764 8.9 
28 .O915 878 9.5 
29 .1059 1082 10.2 
30 .1254 1062 8.4 
31 .1356 (.2732) 1206 8.9 
32° .2578 (.2886) 2395 8.9 
33 .1408 (.2262) 1253 8.9 
Totals 33 5,1078 50,173 9.8 average 


Journat or Comparative NEUROLOGY. 


INGBERT, Dorsal Roots of Spinal Nerves. 93 


Number of Fascicle 


Totals 


© ON Du f WH Nw 


Number of 


Totals 


Oo CONTI OAM PW DN 


Fascicle 


DABLE X. Dh.) 1: 


Number of Nerve 


Number of Nerve 


Area of Fascicle Fibers in Each Fibers in Thousands 
in sq. mm. Fascicle per sq. mm. 

-1579 1455 9-2 

.2267 2208 9.7 

.2895 2653 9.2 

-1155 1063 9.2 

.1778 (.1888) 1627 9.2 

0133 151 jit) 

-1234 (.1497) 1370 Tisi 

-1271 (.1459) I4II ited 

1131 131! HG 

-1744 1762 10.1 

-0964 1045 10.8 

.0729 (.0791) 773 10.6 

-1010 1062 10.5 

1.7890 17,891 10.0 average 

ANB ICE NG pe Dheelt. 
Number of Nerve Number of Nerve 
Area of Fascicle Fibers in Each Fibersin Thousands 
in sq. mm. Fascicle per sq mm. 

.O311 569 17.9 

.0307 418 13.6 

-0334 505 15.1 

.0569 899 15.6 

.0584 721 12.3 

.0627 741 11.8 

-0878 1317 15.0 

.0384 585 15.2 

.0670 1035 15.2 

.0067 88 Hii 

.0207 232 Te 

.0512 608 11.8 

-0720 713 9.9 

-0379 510 13.4 

.0595 848 14.4 

.0331 662 20.0 

.0509 731 14.3 

.0994 1181 11.8 

.O815 1069 eae 

-9742 13,432 13.7 average 


94 JOURNAL OF COMPARATIVE NEUROLOGY. 


Ge PT, NE 


Figs. 12, 15 


INGBERT, Dorsal Roots of Spinal Nerves. 95 


IVeENENS, QUE abel THO 


Number of Nerve Nunaber of Nerve 
Area of Fascicles Fibers in Each Fibers in Thousands 


Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I .0876 1094 12.4 
2 -0466 32 11.4 
3 .0448 614 13.7 
4 20905 1142 12.6 
5 0373 447 12.0 
6 .1672 1946 11.6 
7 -0536 617 11.5 
8 .1078 1236 11.5 
9 .0707 867 12.6 
10 .0841 989 ray 
II .0984 1082 10.9 
12 .0378 479 127 
13 .0472 656 13.8 
svotals)  /13 -9796 11,701 11.9 average 


TAB TE Ey Xone bbe Vi 


Number of Nerve Number of Nerve 
Area of Fascicle Fibers in Each Fibersin Thousands 


Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I .0487 (.0543) 605 WA Al 
2 .0922 1077 11.6 
3 .0584 794 13.6 
4 .0563 698 12.4 
5 .0753 (.0860) 942 12.5 
6 .0984 (.1262) 1231 12.5 
7 .1189 (.1425) 1617 13.6 
8 .1276 1512 11.8 
9 -0992 I1QI 11.9 

10 1281 1708 11.8 


‘Totals 10 .9031 11375 12.5 average 


96 JOURNAL OF COMPARATIVE NEUROLOGY. 


Figs. 14, 15 


INGBERT, Dorsal Roots of Spinal Nerves. 


TNSIOVE DDG Abts AY 


Area of Fascicle 


Number of Fascicle in 


Area 


Number of Fascicle in 


Totals 


O ONAN PW NH 


sq. mm. 


.0926 


-0459 
aye 
.0720 
.0556 
.0756 
.O184 
.0657 
1313 


.6843 


(.1859) 


(.0800) 
(. 1020) 


Number of Nerve 


1198 
605 
1539 
872 
673 
916 
258 
793 
1588 


8352 


TABER XVe Ch. VIL 


of Fascicle 
Sq. mm. 


,O144 
.0243 
.0565 
.0201 
.0252 
.0306 
-0173 
.0226 
LOU52 
.0288 
.0082 
.0633 
+0439 
.O812 


-1377 
-5893 


(.0305) 
(.0958) 


(.0359) 


Number of Nerve 


Fascicle 


248 
294 
684 
244 
306 
269 
146 
402 
249 
432 
IgI 
793 
537 
888 


1472 
7155 


97 


Number of Nerve 
Fibers in Each Fibers in Thousands 


Fascicle per sq. mm. 


12.8 
13.1 
12-1 
ria, f 
201 
12 
14.0 
10.7 
12.0 


12.2 average 


Number of Nerve 
Fibers in Each Fibersin Thousands 
per sq. mm. 


12.1 average 


OURNAL OF COMPARATIVE NEUROLOGY. 


InGBERT, Dorsal Roots of Spinal Nerves. 99 


TABLE XVI. 


Area of Fascicle 


Anan WAL 


Number of Nerve 


Number of Nerve 


Fibers in Each Fibers in Thousands 


Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I -O154 208 13.5 
2 .0549 725 13.2 
3 .0265 482 17.9 
4 .0645 760 Wai 
5 -0325 539 16.5 
6 .0253 306 12.0 
7 +0339 392 11.6 
8 .O161 286 7/9] 
9 +0536 747 13-9 
IO .1073 1265 11.7 
II .0388 718 18.5 
2 -1914 1871 9.8 
13 -0420 756 18.0 
14 -1304 (.1500) 1526 7 
15 -1490 (.2391) 1744 11.7 
Totals 15 .9819 12,325 12.5 average 
TABLE, XViliey ihe Valle 
Number of Nerve Number of Nerve 
Area of Fascicle Fibers in Each Fibers in Thousands 
Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I .0462 497 10.7 
2 .0480 548 ie 
B -0445 561 12.6 
4 .0350 458 13-0 
5 .0223 330 14.7 
6 .0494 507 10.2 
7 .0251 327 13.0 
8 -O132 183 13.9 
9 .0396 542 Ac 
10 -0300 (.0458) 373 12.4 
II .0251 307 22 
12 .0686 (.1304) 837 122 
13 -0594 (.0825) 726 12.2 
14 .0762 (.1047) 937 1228 
15 -0405 577 14.3 
16 .O19S 285 14.3 
17 .0673 988 14.3 
Totals 17 .7102 8983 12.6 average 


OURNAL OF COMPARATIVE NEUROLOGY. 


INGBERT, Dorsal Roots of Spinal Nerves. 101 


PAV BIE eXOVeDII (5 Mae OX 


Number of Nerve 


Number of Nerve 


Area of Fascicle Fibers in Each Fibers in Thousands 


per sq. mm. 


12.2 
122 
riwige: 
11.4 
15.1 
TG.a 
12.2 
11.6 
liitgle 
12:2 
10.6 
12-2 
12.2 
L202 
252 


12.l average 


Number of Nerve 


per sq. mm. 


12.4 
12.4 
12.4 
137 
14.4 
12.4 
12.4 
1L.7 
11.9 
11.9 


11.9 
202 


12.2 
12.2 
12.2 
12.2 


Number of Fascicle in sq. mm, Fascicle 
I -0361 (.0494) 441 
2 0477 (.0684) 573 
3 .0407 528 
4 -0614 702 
5 .0264 399 
6 -0368 555 
7 .0221 (.0348) 280 
8 +0305 355 
9 .0902 1043 
10 -0798 (.0916) 973 
II .0250 265 
12 0405 (.0542) 494 
13 .0436 (.1058) 532 
14 .0362 (.0659) 442 
15 -0472 (.0516) 581 
Totals 15 .6702 8163 
(ASB Bye XG VE hee oe: 

Number of Nerve 
Area of Fascicle Fibers in Each Fibers in Thousands 

Number of Fascicle in sq. mm. Fascicle 
I .0764 952 
2 .0517 (.0661) 640 
3 .0660 (.0819) 819 
4 .0260 358 
5 .0268 387 
6 .0523 (.0620) 649 
7 .0424 (.0487) 526 
8 .0588 686 
9 -0552 (-0930) 657 
if) .0697 833 
II .0635 (.0786) 756 
12 .0465 568 
13 -0273 (.0364) 334 
14 -0554 676 
15 .0609 (.0679) 743 
16 .0842 (.1206) 1028 
Totals 16 .8631 10,612 


12.3 average 


oe 


InGBERT, Dorsal Roots of Spinal Nerves. 103 


TABLE XOXs ih; OX: 


Number of Nerve Number of Nerve 
Areaof Fascicle , Fibers in Each Fibers in Thousands 


Number of Fascicle in sq. mm. Fascicle per sq. mm. 

I -0435 562 12.9 
2 -1107 (.1300) 1528 12.9 
3 -0393 479 12.1 
4 0346 449 13.0 
i -0368 515 14.0 
6 .0321 367 11.4 
7 .0367 471 12.8 
8 -0069 81 I1.7 
9 .0474 660 13.9 
10 .0483 614 12.9 
Il .0323 423 13.1 
12 .0283 399 14.1 
13 -OI1g93 (.0257) 253 13.1 
14 .0981 1109 Wie2 
I5 -0542 651 12.0 
16 -0500 731 14.6 
17 .0162 256 15-5 
18 -0352 571 16.2 
19 -0448 (.0560) 510 11.6 
20 -0352 304 11.6 
21 -0335 470 14.0 

Totals, 21 9134 11,403 12.4 average 


TABLE) <XIy Gh. Xi 


Number of Nerve Number of Nerve 


Area of Fascicle Fibers in each Fibers in Thousands 
Number of Fascicle In sq. mm. Fascicle per sq. mm. 
I .0728 719 9.8 
2 -0433 466 10.7 
3 +1341 1652 12.4 
4 -0882 1072 12.1] 
5 -0492 605 12.3 
6 -0416 484 11.6 
7 -0495 696 12.0 
8 .0333 370 10.9 
9 -0578 685 11.9 
10 .O717 653 g.1 
II .0481 425 8.9 
12 -0542 644 11.9 
13 -0619 588 9-5 
14 -0744 668 8.9 
15 -0734 842 11.4 
16 .O710 940 13.2 
17 .0926 1125 123% 
18 .1300 1491 11.4 


Totals. 18 1.2471 14,125 11.4 average 


“ 


InGBERT, Dorsal Roots of Spinal Nerves, 105 


ANID, OMle” Abell 
Number of Nerve Number of Nerve 


Area of Fascicle Fibers in Each Fibers in Thousands 
Number of Fascicle in sq. mm. Fascicle per sq. mm. 

I .0842 876 10.4 
2 .0627 (.0745) 652 10.4 
3 -1429 (.1602) 1491 10.4 
4 -0613 (.0964) 635 10.4 
5 -1008 (*1519) 1045 10.4 
6 -1570 (.1787) 1633 10.4 
7 . 1004 1054. 10.5 
8 -1433 1538 10.6 
9 0832 1032 12.4 
10 SDs 979 8.6 
II -0546 549 10.0 
12 LD 1252 10.7 
13 1243 1245 10.0 
14 -1403 1459 10.4 
15 1071 1350 12.5 
16 .0569 695 14.0 
7 .0467 656 Bob 
18 0543 714 12°33 

Totals 18 1.7507 18,361 10.8 average 

TABLE XXIII. L. II. 
Number of Nerve Number of Nerve 
Areaof Fascicle Fibers in Each Fibersin Thousands 
Number of Fascicle in sq. mm. Fascicle per sq. mm. 

I -0509 661 11.8 
2 -0775 798 10.3 
3 -1C53 962 g.1 
4 -0718 841 ito 
5 -0787 886 II.1 
6 -1026 1146 11.1 
fl -0678 997 14.7 
cs) -1037 984 9-5 
9 .0778 g15 Mh ioy/ 
10 -0969 1046 10.8 
II -0051 768 11,4 
12 - 1004 1146 11.4 
13 -0838 (.0872) 937 11.4 
14 -0513 (.0590) 585 11.4 
I5 .0873 1089 12.4 
16 .0240 (.0338) 274. 11.4 
17 -0298 (.0397) 335 11.4 
18 -1007 (.1036) 1148 12.4 
19 .0446 692 15.5 
20 -1174 998 8.5 
21 -0609 779 12.7 
22 .0400 514 12.8 
23 -0840 791 9.4 
24 -1010 983 9.7 
25 1171 1006 8.6 
26 .1055 983 9.3 
27 1713 1373 8.0 


Totals 27 2.2172 23,640 10.6 average 


La 


InGBERT, Dorsal Roots of Spinal Nerves. 107 


TABLE XXIV. L. III. 


Number of Nerve Number of Nerve 
Area of Fascicle Fibers in Each Fibers in Thousands 


Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I -1773 2354 13.2 
2 .0324 455 14.0 
3 -1397 154! 11.0 
4 .0301 285 9.4 
5 BI: 1023 9.0 
6 .1376 1624 11.8 
7 .0872 1250 14.3 
8 .O84I 1146 13.6 
9 .0387 509 Tee 

10 mg2u 1644 12.4 
II .0885 965 . 10.9 
12 .0983 1177 12.9 
13 .1304 1584 12.1 
14 .0893 987 11.0 
15 .0884 1059 It.9 
16 -1145 1214 10.6 
17 .0566 793 14.0 
18 .0848 1162 1337) 
19 .0846 1253 14.7 
20 .0964 IO14 10.5 
21 .0958 943 9.8 
22 .0622 997 12.8 
23 .0519 719 13.9 
24 .1767 2446 13 8 
25 .0720 g1o 12.6 
26 0336 538 16.0 
27 -0974 1273 13.0 
28 -0370 463 1225 


Totals 28 2.5309 31-328 12.3 average 


108 


JOURNAL OF COMPARATIVE NEUROLOGY, 


InGBERT, Dorsal Roots of Spinal Nerves. 109 


TABLE XXV. L. IV. 


Number of Nerve Number of Nerve 
Area of Fascicle Fibers in Each Fibers in Thousands 


Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I .0405 707 17.0 
2 .0780 1131 14.5 
5} -0309 725 23-4 
4 .0787 968 12,3 
5 0499 696 13-7 
6 .0420 649 15.4 
7 .0475 887 18.6 
8 .0315 521 16.1 
9 -0572 959 16.7 

10 .0474 824 17.4 
II .0643 1142 17.6 
12 . 1102 1311 11.8 
13 .1184 1389 ni o7/ 
14 .0294 491 16.7 
15 .0765 992 12.9 
16 .0875 997 F138) 
17 .0567 761 13.4 
18 .1792 1537 8.5 
19 .1523 1517 9.9 
20 .0677 993 14.8 
21 .0866 1210 13.9 
22 .0656 992 14.1 
23 .O19 426 21.4 
24 .057 942 14.2 
25 .0485 640 W252 
26 .0467 621 1ge2 
27 .0830 879 10.6 
28 .0826 1056 12.7 
29 .0770 gIo 11.8 
30 .0843 1007 11.9 
31 .1209 1633 13.5 
32 1144 1725 15.0 
33 .0485 595 22 
34 0577 804 13.9 
35 -0505 959 18.9 
36 .0564 865 B53 
37 .0690 1063 15-4 
38 1702 2228 13.0 
39 -1491 1901 12.8 


Totals 39 2,9340 39,653 13.5 average 


INGBERT, Dorsal Roots of Spinal Nerves. III 


TABLE XXVI. L.V. 


Number of Nerve Number of Nerve 


Area of Fascicle Fibers in Each Fibers in Thousands 
Number of Fascicle in sq. mm. Fascicle per sq. mm. 

I -1550 1903 12.2 
2 .0925 1301 14.0 
3 -0976 1280 tion 
4 siuirgis 1431 12.5 

5 -0767 1206 iS 597/ 
6 .0542 835 15-4 

7 -1139 1681 14.7 
8 -0516 852 16.5 
9 .0610 875 14.3 
IO .0810 1218 15.0 
II .0891 1379 15-4 
12 .0257 726 28.2 
13 .0895 1273 14.2 
14 20553 914 16.5 
15 -0443 739 16.6 
16 -1306 1607 T2a3 
17 0103 332 Beg 
18 -1595 1622 10.1 
19 -0125 483 17.8 
20 .1138 1280 11.2 
21 .0827 1058 127) 
22 -0604 687 1160) 
23 .1296 1338 10.3 
24 .1305 1504 11.4 
25 - 1086 1304 12.0 
26 .0758 1394 14.3 
27 -O181 326 18.0 
28 .1050 1467 13.8 
29 2525 "2490 9.9 
30 .0865 966 Il. 
31 .1189 1760 14.8 
32 .0346 561 14.1 
33 .0233 383 16.4 
34 .0869 1363 ay 
35 .0470 763 16.2 
36 -1189 1639 13.8 
37 ,0951 1193 an 


Totals 37 3,2020 43,128 13.4 average 


112} = JouRNAL oF ComPaRATIVE NEUROLOGY. 


INGBERT, Dorsal Roots of Spinal Nerves. 113 


ADEE ID, ROOQVINS SHI 


Number of Nerve Number .of Nerve 
Area of Fascicle Fibers in Each Fibersin Thousands 


Number of Fascicle in sq. mm. Fascicle per sq. mm. 
I -2955 3665 12.4 
2 -1259 1606 W2e7) 
3 -1824 2816 15-4 
4 -0939 1680 Yes 
5 -0430 724 16.8 
6 -0924 1292 13.9 
7 -1423 1839 12.9 
8 -0530 993 18.7 
9 -1114 1622 PLANS 

10 0946 1496 15.8 
Ii -1266 1331 10.5 
12 .0899 1187 13.2 
13 -1075 1293 12.0 
14 .2292 2980 13.0 
15 -0976 1552 15.8 
16 0326 821 25.1 
17 .O212 492 2B) 
18 1212 1526 12.5 
19 -0499 936 18.7 
20 -0908 1461 16.0 
21 -1367 1977 14.4 
22 2341 2916 12.4 
23 .1540 2079 1315 
24 -0682 1020 14.2 
25 -2603 3178 12.2 
26 -IOII 1279 12.6 
27 - 1107 1482 ges 
28 -1734 2218 L257 


Totals 28 304394 * 47,461 13.8 average 


j aa: a 
ae 
Aly kee 
es 
ae 


INGBERT, Dorsal Roots of Spinal Nerves. 115 


AEAN BIOS, MOVIN Se WILE 


Number of Nerve Number of Nerve 


Area of Fascicle Fibers in Each Fibers in Thousands 
Number of Fascicle in sq. mm. Fascicle per sq. mm. 

I -1135 1524 15.6 
2 -1651 2298 13.9 
3 .0270 382 14.1 
4 .0871 1373 15-7 
5 -1356 1251 Q.2 
6 .0849 1052 12.3 
7 .0312 595 19.0 
8 .0652 866 12303) 
9 .0315 631 20.0 
10 .O861 1197 13.9 
II .1189 1652 139 
12 .0367 559 fils 
13 .0935 1128 12.0 
14 = .0728 ia 15.4 
15 .2592 3079 11.9 
16 O27, 2274 13.9 
17 .3504 4562 13.0 

Totals 17 1.9214 25,545 13.3 average 

TABLE XXiIxX. Suu 
Number of Nerve Number of Nerve 
Area of Fascicle Fibers in Each Fibers in Thousands 
Number of Fascicle in sq. mm. Fascicle ; per sq. mm. 

I 2a 2027 16.4 
2 .1672 2490 14.8 
3 . 1092 1565 14.3 
4 . 1099 1508 14.2 
5 . 1696 2205 13.0 
6 .0916 1452 15.8 
7 BIR 1815 13.6 
8 .1029 1389 tals 
9 - 1091 2185 20.0 
10 .0594 626 10.5 


Totals 10 Tey 5a 17,322 14.9 average 


L116 


JOURNAL OF COMPARATIVE NEUROLOGY. 


an 


een 
ays 


ay 


Ce ame 


Figs. sO, 31, 52 


InGBERT, Dorsal Roots of Spinal Nerves. U7 


TA BUEN SuLvs 


Number of Nerve Number of Nerve 


Area of Fascicle Fibers in Each Fibers in Thousands 
Number of Fascicle in sq. mm. Fascicle per sq. mm. 
.0588 794 15.9 
2 .0488 997 26.5 
3 .0366 I10L 29.0 
4 .0434 I161 30.0 
5 .04.40 1392 20.4 
6 -0330 958 ae 
OT) .1368 2177 210 
Totals 7 .4014 8580 21.3 average 
TABOR XOXO Save 
Number of Nerve Number of Nerve 
Area of Fascicle Fibers in Each  Fibersin Thousands 
Number of Fascicle in Sq. mm. Fascicle per sq. mm. 
I -0938 1719 “18.3 
2 .0220 507 23.0 
totals 2 1258 2226 Ig.I average 
ABER OSX COCe Ic 
Number of Nerve Number of Nerve 
Area of Fascicle Fibers in Each Fibers in Thousands 
Number of Fascicle in sq. mm. Fascicle per sq. mm. 


I .0324 761 23.5 


118 JOURNAL OF COMPARATIVE NEUROLOGY. 


ERRATA. 
Page 54, line 15 from the bottom, for 7. Rosenthal, read D. Rosenthal. 
Page 59, lines 11, 12 and 13, for mm. read mm’. 
Page 60, line 12 from bottom, for 56.39, read 56.319. 


Page 63, line 14 for Since the magnification used, read Since the magnt- 
fication of the objective used. 


Ga 


INGBERT, Dorsal Roots of Spinal Nerves. 119 


BIBLIOGRAPHY. 
SALZER, F., 1880. Svtzungsb.d. K. Akad. ad. Wissensch. zu Wien, Math. 
Naturw. Classe, Wien. Bd. LXXXI, Abth. III. pp. 7-23. 
KRAUSE, W.. 1880. Arch. f. Ophth. Bd. XXVI, Abth. II, p. 102. 
DoNALDSON, H. H., and Botton, T. L., 1891. Am. J. Psychol., Vol. 
IV, pp. 224-229. 
ROSENTHAL, D., 1845. De numero atque mensura microscopica fibrillarum 


elementarium systematis cerebro-spinalis symbolae. Dissertatio. Vvratz- 
slaviae. 


TERGAST, P., 1872. Arch. f. mikr. Anat., Bonn. Bd. 1X, Heft I, pp. 
26-46. ' 
KuuHntT, H., 1879. Arch. f. Ophth. Bd. XXV Abth. III, p. 179. 


Krause, W., 1876. Anatomie, Hannover, p. 165. 
a “©. 1880. Arch. f. Ophth. Bd. 26, Abth. II, p.-102. 


“e ce ee oe es es ce ee 


KO6OLLIKER A, 1850. Microscopische Anatomie. Lepzig, p. 433. 


STILLING, B., 1859. Neue Untersuchungen itiber den Bau des Riicken- 
mlanksseniCasses.\ ibe sAo. 


STILLING, B., 1859. Atlas. Cassel. Taf. XVIEI. 
Donatpson, H. H., 1894. /. Aforphol., Boston, Vol. IX, No. 1, p. 154. 
Dona.pson, H. H. and Davis, D. J., 1903. 7. Comp. Neur., Vol. XIII, 
No. 1. 
STILLING, B., 1859. Same as No. II, p. 600. 

at GE ot coo) Sppa67so 08nd. O70: 
Bipper, F. H., and VOLKMANN, A. W., 1842. Die Selbstindigkeit des 
sympathischen Nervensystems, Lezpzig. 
KO6LLIKER, A., 1850. Handbuch der Gewebelehre, 3 Auflage. Lezpzzg, 
p. 285. 
DUCHENNE, 1863-4. Bull. Acad. de med., Paris, XX1X, 1008-1012. 
SIEMERLING, E., 1886. ' Veur. Centralbl., Leipzig, Vol. IV. 

ce «© 1887. Anat. Unters. iib. d. Menschl. Riickenmarkswurz- 
eln. Berlin. 


BELL, CHARLES, 1811. Idea of a New Anatomy of the Brain. London. 
(Ms. copy). ‘ 


v. LENHOSSEK, 1890. Anat. Anz., Jena, Vol. IV, p. 360. 


CAJAL, S. RAMON Y, 1890. Anat. Anz., Jena, Vol. IV, p. 112. 
ce (3 1893. sé 3 6é 6c Wale 
VAN GEHUCHTEN, A., 1893. ‘ es Cee Ceeen Ville 


RETz1uUSs, GusTAv, 1892. Biol. Untersuch., Stockholm. n. F. Bd. V. 
MartTIN, Is., 1895. La Cellule. Vol. XI, p. 55. 

STEINACH, E., 1893. Lotos. Prag. Bd. XIV. 

STEINACH E., 1898. Pfliiger’s Arch. f. Phystol., Leipzig. Bd. UXXI. 


+ 


JoURNAL OF COMPARATIVE NEUROLOGY. 


STEINACH, E., and WIENER, H., 1895. Pfiiger’s Arch., Vol. LX, p. 593. 
SmiTH, H., 1897. /. Physiol., London, Vol. XXI, p. Iot. 


VEJAS, P., 1883. Beit. z. Anat. u. Physiol. d. Spinale Gangl., Inaug. Diss. 
Munich. 


JosErH, M., 1887. Archiv f. Physiol., p. 296. 
KAHLER, O., 1884. Prag. med. Wehnschr. p. 302. 


SINGER, J., and M@tnzer, E., 1890. Denkschr. d. k. Akad. d. Wissench. 
Wien. Bd. LVI, p. 570. 


SHERRINGTON, C. S., 1894. /. Physiol., London, Vol. XVII, p. 211. 
SHER RINGRONG)| C2) So, moO} aon mcs GG XOX pp 2kos 


VAN GEHUCHTEN, A., 1895. Bull. Acad. roy. d. sc. de Belg. Brux. 
Gapsrl, G., 1896. Arch. ttal. de biol., Turin, p. 115-119. 


STRICKER, S., 1876. Sttzungs d.k. Akad. d. Wissensch. Wien, LXX1V 
Abth. III. 


Bonuzzt, P., 1885. Med. Jahrb., Wien, 473. 


Bonuzzi, P., 1886-87. Adtti. r. Accad, med. di Roma. 2.s., Il, 295-337. 
Abstract in Centralbl. f. Physiol., Leipzig u. Ween. Bad. I. 


GarRTNER, G., 1889. Wren. klin. Wehuschr. p. 980. 
Morat, J. P., 1892. Arch. d. Phystol. p. 689. 


HASTERLIK, P., and BIEDL, A., 1893. Wien. klin. Wehnschr. Bd. VI, No. 
3. Abstract in Centralbl. f. Phystol., Letpzig u. Wien, 1893. p. 249. 


WERZILOFF, N. M., 1896. Centralbl. f. Physiol. Leipzig u. Wien, p. 194. 


Bay.iss, W, M., 1900. /. Physiol., London. Vol. XXVI, p. 172. 
BAYLIss, W. M., Igo2. ‘S 3 oC Vol. XXVIII, No. 4. 


Roux, JEAN-CuH., 1900. Les lesions du systéme grand sympathique dans 
le tabes, etc. fares. 


SHERRINGTON, C. S., 1894-95. /. Physiol., London, XVII, p. 211. 
Harpesry, I., 1899. J. Comp. Neur., Vol. IX, p. 65. 
Harpvesty, I., 1900. J. Comp. Neur., Vol. X, p. 323. 


DuNN, ELIZABETH H., Igoo. * §S SOX peezlge 
ce “ec 6c 1902. “é se ce >.41 IP p: 297. 


ON THE PHYLOGENY AND MORPHOLOGICAL POSI- 
DIONIOE THE TERMINAL BUDS OF FISHES:* 


By C. Jupson Herrick. 


These curious sense organs occurring in the skin of certain 
fishes have been a source of perplexity and controversy among 
morphologists since their discovery by Leypic in 1851. They 
are found freely scattered over the surface of the skin of the 
head and trunk in certain teleostean and ganoidean fishes, par- 
ticularly in exposed situations, and hence by many observers 
have been regarded as tactile organs. Indeed, MERKEL in 
1880, having failed altogether to find the proper tactile nerves 
in the skin of fishes belonging to what we now designate as the 
general cutaneous system, supposed that these organs, together 
with the neuromasts or organs of the lateral line system, were 
developed in compensation for the absence of the typical free 
tactile nerve endings of the other vertebrates. This we now 
know to be wide of the mark, for all fishes which possess either or 
both of the systems of special sense organs mentioned above 
also possess in the same cutaneous areas an abundant general 
cutaneous nerve supply for tactile sensation. 

We must distinguish at the outset three distinct types of 
sensory nerve endings in the skin of fishes and then inquire in- 
to their respective morphological rank; viz., (1) the general 
cutaneous nerve termini, (2) the neuromasts, or organs of the 
acustico-lateralis system of sense organs, (3) the terminal buds, 
or end-buds. 

(1) The first type comprises the ordinary tactile nerves, 
making up the greater part of the spinal dorsal roots, but rep- 
resented in certain ones only of the cranial nerves. They are 


1 Studies from the Neurological Laboratory of Denison University, No. 
XVII. 


122 JOURNAL OF COMPARATIVE NEUROLOGY. 


free nerve endings in the skin without specialized sense organs 
and may be regarded as in all probability the most primitive 
type of sensory ending. 

(2) Neuromasts. This system of sense organs includes 
_the lateral canal organs, pit organs, ampullae and all other 
specialized organs associated with the lateral line canals and in- 
nervated by lateralis nerves, together with the sense organs of 
the internal ear of like phylogenetic origin and innervation. 
These organs clearly belong to asingle system and the evidence 
thus far accummulated favors the inference that the system asa 
whole has been derived phylogenetically from the general cuta- 
neous system of nerves. This is not the place to give the de- 
tailed proof of this, but the trend of the current argument may 
be summarized as follows under two heads: 

(a) the sense organs themselves are characterized in their 
adult and highly functional condition by the presence of hair 
cells differentiated from indifferent supporting cells and extend- 
ing only part way through the epithelium of the sense organ. 
These hair cells may not occur in the earliest embryological 
stages of development of the organs, nor in states of functional 
and structural degradation. The former point accounts for 
the statements of several authors to the effect that the lateral 
line organs pass through a developmental stage which is similar 
to the adult structure of the terminal buds, and the latter point 
for the confusion which has arisen in the minds of many 
authors, giving rise to the belief that there is no clearly marked 
structural difference between neuromasts of the lateral line 
series and the terminal buds. Nevertheless, the distinction 
drawn by ScuuLzE and MerKEL between the neuromasts and 
terminal buds, the former possessing the hair cells and the 
latter not, stands as an essential criterion in all cases where the 
organs are well developed. Now these hair cells resemble 
somewhat in structure and probably also in mode of function 
the tactile hairs of certain aquatic invertebrates and in fact are 
probably phylogetically derived from them. ParKER (’03) has 
shown clearly the probable line of differentiation of these 
structures from tactile organs for the perception of mass move- 


Herrick, Zevminal Buds of Fishes. » 123 


ment of the water to organs of equilibration and hearing. We 
therefore have structural and physiological evidence that the 
sense organs of the acustico-lateralis system have been derived 
from tactile organs. 

(b) Now turning to the central connections of the acustico- 
lateralis nerves within the brain, we have a quite independent 
line of evidence. It has long been known that these nerves all 
end in the tuberculum acusticum and associated centers, and it 
has recently been shown, particularly by JOHNSTON in an im- 
portant series of papers, that these centers are all derivatives of 
the general cutaneous or tactile centers of the brain. While 
my own studies on the teleosts have shown that in these higher 
fishes the specialization of the acustico-lateralis centers has 
progressed so far as to obscure the primary relations, JoHN- 
STON’s observations on the lower fishes leave no room for doubt 
that the history is as stated above. 

From the concurrent history of both central and peripheral 
relations we are therefore justified in assuming that the neuro- 
masts and their associated nervous apparatus have been de- 
rived from the general cutaneous or tactile system of nerves. 

(3) The terminal buds (end-buds, beaker organs, etc.) are 
by no means on 30 secure a morphological foundation. It is, 
however, now definitely known that they resemble in every es- 
sential respect the taste buds within the mouth cavity. The 
specific sensory cells are not hair cells, but, like the indifferent 
supporting cells, they run through the entire thickness of the 
sensory epithelium from external to internal limiting membrane. 
They may terminate distally in a short stiff bristle, but never in 
the tuft of long hairs characteristic of hair cells. The organs 
usually rest on a raised papilla of dermis, and finally in every 
case where the innervation is accurately known they are sup- 
plied by communis nerves, which have nothing in common 
either centrally or peripherally with the general cutaneous or 
acustico-lateralis systems of nerves. On the other hand, these 
nerves always terminate within the brain in a single center 
(bilobed in some fishes), the vagal lobe (plus facial lobe in silur- 


124 JOURNAL OF COMPARATIVE NEUROLOGY. 


oids and cyprinoids), which also receives the typical gustatory 
fibers from the taste buds of the mouth. 

If now it can be shown that the terminal buds differ func- 
tionally, as well as in this thorough-going structural fashion, 
from all other cutaneous sense organs, then it would appear 
that their rank as a separate system of sense organs should re- 
main unchallenged. And _ such in fact is the attitude of most 
of the recent students of this question, even without a rigid de- 
monstration of the functional relations. I am, however, at this 
time in a position to contribute positive facts regarding the 
functions of the terminal buds, and since one recent author of 
note has within the year made a very forcible plea for a return 
to the older standpoint of Leypie and others that terminal buds 
and neuromasts are genetically related, I am moved to review 
the whole question again. 

But first we must examine more thoroughly the central 
nervous connections of the terminal buds, since the morphologi- 
cal argument really hinges upon this point. I repeat, that the 
terminal buds of the outer skin of fishes are known to be inner- 
vated by a-system of nerves which is enterely distinct throughout 
from either the tactile or lateralis nerves. There are no accu- 
rate observations which contradict this conclusion, while the 
positive evidence is quite decisive. The most convincing chap- 
ter in the elucidation of this problem is, I think, my own exam- 
ination of the innervation of the cutaneous organs in Ameiurus 
(Herrick, 01), which was undertaken primarily to test this very 
question. The cyprinoids would give an equally decisive an- 
swer to the problem, as I know from personal observation (as 
yet unpublished), though perhaps here the evidence would not 
be quite so easy to read. | 

The nerves from these terminal buds, no matter where 
they are located on the body, always reach the central nervous 
system through cither the X, LX or VII pairs of cranial nerves, 
and so far as my personal observation goes, practically all come 
in by the VII nerve, though all three of these nerves in all 
types receive gustatory fibers from taste buds within the mouth. 
Both of these classes of fibers, together with a large number of 


Herrick, Zerminal Buds of Fishes. 125 


unspecialized visceral sensory fibers must be treated together 
within the brain, since they are so intimately intermingled that 
analysis has so far proven impossible. Accordingly, they are 
termed collectively the ‘‘communis system”’ of cranial nerve 
fibers. Peripherally this system is clearly divisible into two 
components, (1) the unspecialized visceral sensory which is 
doubtless much more ancient phylogenetically, and (2) the 
specialized, which for reasons to appear we may now term the 
gustatory component. This component may be divided topo- 
graphically into two divisions, one for taste buds within the 
mouth, and one for terminal buds in the outer skin. It is the 
latter division only, of course, with which we are here con- 
cerned. 

In fishes in which the terminal buds are not exceedingly 
numerous, as in the cod, their nerve fibers, though they enter 
the brain by the facial nerve, all turn back in the fasciculus 
communis to find their terminal nucleus in the vagal lobe, along 
with all other communis nerves of the body. But in the two 
groups of teleostean fishes in which these organs are most 
abundant (viz., the siluroids and cyprinoids) their nerve fibers 
find their terminal nucleus at the cephalic end of the fasc. com- 
munis in a special center, the lobus facialis. In both of these 
groups. all of the terminal buds are innervated from the genicu- 
late ganglion of the facialis root. Of this I am confident in the 
siluroids (HERRICK, ’o1). As to the cyprinoids, a rather 
cursory examination of sections of Carassius convinces me that 
the same is true, though my study of this type is not by any 
means exhaustive. 

The facial lobe of the siluroids was formerly known as the 
lobus trigemini, that of the cyprinoids as the lobus impar, in 
the former case the structure being paired, in the latter un- 
paired by the fusion in the dorso-median line of the lobes of 
the two sides. That the facial lobes in these two groups of 
fishes are really merely differentiated parts of the vagal lobes is 
manifest, not only from the nature of their peripheral connec- 
tions as indicated above, but still more evidently from their. in- 
ternal structure and secondary fiber connections. I can speak 


126 JOURNAL OF COMPARATIVE NEUROLOGY. 


with confidence on the latter point, because I have now in an 
advanced state of preparation a study of their connections and 
internal organization based on an extensive series of sections by 
several standard histological methods, including the methods of 
WEIGERT and GOLGI. 

It will not be necessary for me at this time to report these 
histological observations further than to say that they abund- 
antly confirm the conclusions to which JonnsTon was led in his 
studies of Acipenser (’01) and Petromyzon (’o2), that the 
primary and secondary connections of the communis system 
within the brain are absolutely distinct throughout from those 
of either the general cutaneous or the acustico-lateralis systems. 
This I regard as a matter of the highest importance to any es- 
timate of the morphological interpretation of the sense organs 
innervated from these various centers 

We may now turn to the consideration of the function of 
the terminal buds. This is a matter to which I have devoted 
considerable attention during the past year and with results 
which I think may be regarded as decisively answering the 
question. A preliminary account of these experiments was 
presented to the American Association for the Advancement of 
Science at the Pittsburg meeting (Abstract appearing in Sczence, 
HERRICK, ’02) and the final report upon the problem is,now in 
press in the Bulletin of the U. S. Fish Commission at Wash- 
ington. From that report I excerpt the following summary of 
results. 

The entire cutaneous surface of Ameiurus is known to be 
supplied with terminal buds, and particularly the barblets. In 
the gadoid fishes terminal buds are known to be present on the 
barblet and on the free filiform rays of the pelvic and dorsal 
fins. Accordingly I chose as the chief subjects of investigation 
Ameiurus nebulosus, the common ‘horned pout,’ Microgadus 
tomcod, the ‘tom cod,’ and Urophycis tenuis, the ‘hake,’ to- 
gether with a number of fishes such as Prionotus and Opsanus 
in which microscopical examination has shown that terminal 
buds are absent from the outer skin (MorriLi, ’95 and 
CLAPP, ’99). 


i Di 


Herrick, Zerminal Buds of Fishes. 127 


I may say at once that the fishes known to lack terminal 
buds and the corresponding cutaneous branches of the commu- 
nis system of nerves in all cases failed to give any response to 
any kind of gustatory stimulus applied to the outer skin, in 
marked contrast to the other types experimented upon. 

The methods of experimentation upon the siluroid and 
gadoid fishes in all cases were exceedingly simple, the attempt 
being to approach as nearly to the normal feeding habits of the 
species in question as possible. In particular, care was taken 
to select only such gustatory stimuli as the fishes were already 
familiar with and for which there were ready formed well de- 
fined reflex paths. 

The type of reaction to gustatory stimulation of the ter- 
minal buds was found to be practically constant whether the 
organs stimulated were situated on the dorsal fin (Urophycis), 
on the ventral fin (Urophycis and Microgadus), on the barblets 
(Microgadus and Ameiurus) or on the general body surface 
(Ameiurus). The participation of visual sensations in the reflex 
act evoked by the stimulus was excluded in certain of the cases 
by the conditions of the experiment; e. g. in Ameiurus the 
stimulus could be presented to the terminal buds of the body 
or a barblet when the head was concealed under leaves or 
debris on the bottom of the tank. I did not think it necessary 
to blind any of my fishes because the experiments were suffh- 
ciently clear without it, and besides BATESON (’90) has already 
tried that experiment on certain European gadoid fishes, show- 
ing that the reactions now under consideration are not affected 
by that operation. The participation of the sense of smell was 
excluded in the case of the tom cod by the destruction of the 
olfactory organ, with no noticeable resultant modification in the 
reflexes in question. 

The problem is then narrowed down to the differentiation 
between the senses of taste and touch in my experiments, and 
this I think can easily be accomplished. My most common 
mode of procedure was to touch the parts of the body where 
terminal buds are known to be abundant with sapid substances 
such as bits of meat, clam, etc., on the end of along slender _ 


ob by : A 
“0 


Aired qs 

q Ly i > “A WS i 
ee se \o 
Lo {LIBRARY | a 
\Z Om = 


128 JOURNAL OF COMPARATIVE NEUROLOGY. 


wire, when possible so arranging the experiment that the fish 
could not see the point of contact. The fish immediately 
turned witha characteristic movementand snapped up the morsel. 
But the areas in question, particularly the barblets and free fili- 
form fin rays, are known to be very richly supplied with gen- 
eral cutaneous nerves in addition to the communis nerves for 
the terminal buds; in other words they have a very rich tactile 
innervation and doubtless are very sensitive to touch. Was 
the reaction then essentially tactile or gustatory or both? To 
test this point I replaced on the end of the wire the customary 
bit of meat with a wisp of cotton wool or a piece of colorless 
and tasteless gelatin which had been previously softened 
up in cold water. The fishes sometimes would take it at 
the first contact, though they would rarely repeat it and 
soon ceased to notice the cotton or gelatin at all. If now the 
cotton were soaked in meat juice and the contact repeated, the 
fish always instantly reacted exactly as he did to meat in the 
first instance and no amount of training would suffice to cause 
him to discontinue the reaction to the cotton when dipped in 
filtered meat juice. 

In brief, the experiments which were hundreds of times 
repeated in a great variety of forms, show that the fishes nor- 
mally react to both tactile and gustatory stimuli upon the parts 
of the body in question ; but by training they can be induced to 
discriminate between the two classes of stimuli, ignoring the 
simple tactile stimulus, but reacting to this plus the gustatory. 
I varied the experiment, among other ways, by the use of a 
fine-pointed pipette, directing a jet of water against the fish 
and then similarly applying a jet of filtered meat juice. In'the 
former case the jet was ignored or avoided; in the latter it was 
always eagerly sought, the reaction being similar to that pro- 
duced by a contact with meat, though more pronounced. 

It may be regarded as established that fishes which possess 
terminal buds in the outer skin taste by means of these organs 
and habitually find their food by their means, while fishes 
which lack these organs in the skin have the sense of taste con- 
fined to the mouth. The delicacy of the sense of taste in the 


—d 
— a 


Herrick, Zerminal Buds of Fishes. 129 


skin is directly proportional to the number of terminal buds in 
the areas in question. 

Numerous unrelated types of bony fishes from the siluroids 
to the gadoids which possess terminai buds have developed 
specially modified organs to carry the buds and increase their 
efficiency. These organs may take the form of barblets or of 
free filiform fin rays. The free rays of the pelvic and dorsal 
fins of gadoid fishes are thus explained, and indeed this is pos- 
sibly the motive for the migration into the jugular position of 
the pelvic fins of the gadoids. In all cases where terminal 
buds are found on barblets or filiform fin rays gustatory nerves 
belonging to the communis system are distributed to them. 

The fishes in which the cutaneous terminal buds are most 
highly developed are in general bottom feeders of rather slug- 
gish habit and in some cases they are nocturnal feeders. The 
high development of this sense is compensated for in some fishes 
by the reduction of others. The visual power of the fishes is 
especially apt to suffer degradation. This degradation may be 
organic, a positive degeneration of the visual apparatus, as in 
Ameiurus, or it may be merely functional. In the latter case, 
though the organs of vision are not necessarily modified, these 
organs are not actually used in procuring food, the fish being 
unable to effect visual reflexes toward food substances or to cor- 
relate visual stimuli with the movements necessary to react to- 
ward food substances. The fish may be perfectly able to effect 
other visual reflexes, such as avoiding enemies, but is appar- 
ently unable to understand the significance of food when per- 
ceived by the sense of sight only. This particular central re- 
flex path has never been developed, or has atrophied from dis- 
use. Nature has here effected for the species something simi- 
lar to what is accomplished in individual men occasionally by 
disease, in the production of certain aphasias. 

This study has been directed primarily toward the solution 
of a simple physiological problem; but in a purely incidental 
way some points of interest to comparative psychology have 
come up. We have seen that in the cat fish, hake and tom 
cod the reflex of seizing the food is normally set off by a com- 


130 JOURNAL OF COMPARATIVE NEUROLOGY. 


bined stimulus of tactile and gustatory end-organs. At first 
the fish may react similarly to a pure tactile stimulus and the 
tactile plus the gustatory. After a brief training, however, he 
acquires the ability to discriminate between the former, which is 
never followed by satisfaction, and the latter, which is followed 
by the pleasure of feeding. Clearly the fish learns by experi- 
ence. We find also some differences between the different spe- 
cies of fishes in this respect, depending on the relative impor- 
tance of the tactile and gustatory elements of the sensation com- 
plex in the normal reflex life of the fish. 

It would be interesting to inquire the part played by 
memory in these reactions. In the case of Ameiurus, where 
the tactile and gustatory elements of the reflex of seizing food 


can be experimentally isolated by training, it would doubtless 


g, 
be possible to measure quantitatively the duration of the per- 
sistence of this acquired discrimination. I have made no accu- 
rate observations on this point, but can say in general that the 
memory of these fishes seems to be fairly good. (By the term 
memory I do not mean to prejudice the question of the part 
played by consciousness here. The original reaction may be 
largely or wholly an unconscious or automatic response and the 
“memory”? may be an organic memory more closely allied to 
habit). At the beginning of the tests’ with cotton the cat fishes 
generally seized the cotton just as they did the meat. At the 
close of the first day’s experiments they had learned to ignore 
the cotton asa rule, and half an hour after the close of this se- 
ries of tests they still would pay small attention to the cotton ; 
but by the day following, if tested first with meat, they would 
take the cotton for-a few times or would react to it slightly 
during the first few tests, but would learn to let it alone sooner 
than on the first day. But toward the close of the experiments 
after several weeks of practice I rarely got any reaction at all 
with the cotton under any circumstances, even if the fishes had 
not been tested for several days. With the gadoids the num- 
ber of experiments was much smaller and they were continued 
for a shorter time, but I never got so good evidence of memory 
of the discrimination. On successive days the tests were much 


‘ie eS —— — 


a 


7 —— 


Herrick, Zerminal Buds of Frshes. Thee 


alike. The inability of the tom cod to remember to ignore a 
tactile contact which is not followed by satisfaction so long as 
the cat fish remembers a similar discrimination I take to be an 
indication that the tactile element plays a much larger part in 
the reflex complex in the gadoids. | The known distribution of 
the terminal buds favors this view also, for while they are very 
abundant on the barblets and body of the cat fish they are 
rather sparse on the free fins of the gadoids and the general cu- 
taneous nerve supply on the fins of these fishes is greatly in ex- 
cess of the communis nerve supply. 

I noticed also that all of the fishes.that ate freely in 
captivity soon accustomed themselves to novel methods of feed- 
ing and in the case of the cat fishes and the hake especially, as 
soon as I approached their tanks after the experiments had 
been in progress some time, the fishes would rise to the top of 
the tank and eagerly await the expected food. This restless- 
ness became so great with the cat fish that the experiments be- 
came increasingly more difficult and there was evidence that 
vision and possibly smell assumed greater importance after this 
expectation of food had made its appearance. 

From the experiments just summarized we may, I believe, 
conclude that these fishes taste with the terminal buds essen- 
tially as they do with taste buds within the mouth and thus we 
have added the last link in the chain of evidence necessary to 
fix the position of these sense organs, the morphological and 
the physiological evidence giving concurrent testimony to the 
essential isolation of this system from either of the other types 
of cutaneous nerves. Weare not at present able to assert with 
confidence the phylogenetic origin of the gustatory system of 
nerves and sense organs, though the structure of the brain cen- 
_ ters seems to favor the belief that the system has been differen- 
tiated from the general visceral sensory type of nerves. Of this 
it cannot be said that we have adequate proof and experience 
teaches us that, in the absence of evidence, speculations in this 
field are not very profitable. 

As intimated above, the trend of most of the recent work 
on the nervous system of the fishes accords fairly well with the 


132 JOURNAL OF COMPARATIVE NEUROLOGY. 


conclusions here expressed, with one notable exception. Mr. 
EK. P. Atuis (’o1) has recently published an important paper 
on certain features of the sense organs (particularly the am- 
pullae of Lorenzinr) and cranial nerves of Mustelus, which 
takes its departure from our present problem, as indicated from 
the following quotation for the introductory paragraph : 


‘I have long had a very decided impression, opposed to that of 
most workers on the subject, that these ampullary organs must be ge- 
netically related to the terminal buds of ganoids and teleosts rather 
than to the pit organs of those fishes; and I thought that I should 
easily be able to get some positive evidence of this in the general 
course and position of the nerves that innervate them in advanced 
selachian embryos. ‘This positive evidence I have wholly failed to get, 
for the very simple reason that, in the main nerve trunks, I could not 
distinguish in my sections the ampullary fibers from the lateral canal 
ones. Disappointed in this at the very beginning of my investigation, 
I nevertheless decided to quite carefully trace the lateral canals and 
the nerves that innervate them and the ampullae, as far back as my 
sections went, that is, nearly to the level of the first gill slit. Careful con- 
sideration of these observations has fully convinced me, though indi- 


rectly, that the ampullary organs do represent the terminal buds of 


ganoids and teleosts, and not the pit organs.” 


These conclusions have been criticised at some length by 
JOHNSTON (’02) and by myself (this Journal, vol. XII, p. iii) 
but ALrIs (’03) now returns to his original proposition fortified 
by fresh facts from the study of the lateral line system of Polyo- 
don, though the evidence is still all indirect. Fortunately, at this 
time we are able to meet speculation with fact, and, as we have 
seen above, to put the morphology of the terminal buds ona 
secure foundation. 

ALLIS’ argument rests ultimately on two main supports, 
viz.: (1) the supposed homology of the cerebral center from 
which the ampullary organs of selachians are innervated with 
that from which the teleostean terminal buds are supplied ; and 
(2) the supposed similarity of the ampullae themselves and the 
terminal buds. 

(1) On the first point Mr. Artis is able to oppose to the 
observations of all students of selachian nerves that the am- 
pullae are supplied from lateralis branches only the conjecture 
that the twigs for the ampullary organs terminate in the brain 


——=— = 


Herrick, Zerminal Buds of Fishes. 133 


in a special center, the so-called lobus trigemini, and that this 
center is homologous with the structure of that name in Acipen- 
ser and teleosts. 

Now it proves that neither of these assumptions is true, or 
at best they are only partially true. The ‘‘lobus trigemini”’ 
of elasmobranchs is homologous with the part so named in 
Acipenser, but in both cases the part is a derivative of the 
tuberculum acusticum and is related to the peripheral 
nerves from lateral line organs. The work of JOHNSTON 
leaves no room for doubt on this point, and we may adopt his 
designation for this lobe, ‘‘lobus lineae lateralis.”” The so- 
called lobus trigemini of certain teleosts (cyprinoids and silur- 
oids) is now known to bea totally different structure (K1NGs- 
BURY, 97; HERRICK, ’O1), a center for terminal buds, having 
nothing to do with any part of the acustico-lateralis system. It 
is termed by recent writers the lobus facialis, and, in view of 
ALLIS’ perpetuation of the old confusion growing out of a 
false nomenclature, we may well adopt the suggestion of 
Houser (’o1), ‘‘The term “#zgemznal lobe has been so variously 
used that it should be dropped from our nomenclature.” 

We are fortunately no longer shut up to speculation re- 
garding the exact relations of these nerve roots in selachians. 
In the course of a report upon his analysis of the cranial nerves 
of Squalus acanthias, STRONG (’03) writes, ‘“The ramus mandi- 
bularis externus VII is apparently derived practically entirely 
from the more dorsal of the two lateral line roots [of the 
facialis], the ramus buccalis receiving the major part of the re- 
mainder of this root, while the ramus ophthalmicus superficialis 
VII is principally composed of the bulk of the more ventral 
lateral line root. This would apparently negative the view that 
the ampullary organs are modified end-buds and the dorsal root 
-an end-bud root.”’ This destroys completely the foundations of 
A.tts’ labored argument from the structure of the r. ophthalmi- 
cus superficialis. In brief, we may conclude from the works of 
Houser, JOHNSTON, STRONG and others, that the terminal bud 
and lateralis centers in the brains of cyclostomes, selachians and 
ganoids are as distinct as I find them to be in the teleosts and 


134 JOURNAL OF COMPARATIVE NEUROLOGY. 


that the ampullae of selachians are related to the lateralis and 
not to the terminal bud centers. 

(2) The supposed similarity of ampullary organs and ter- 
minal buds we have already touched upon. The fact is that 
these organs do not resemble each other in their well developed 
adult forms in any known case. In embryonic or reduced con- 
ditions the neuromasts may resemble terminal buds on account 
of the absence or reduction of their hair cells. But even in 
these cases the innervation in all instances where it is accurately 
known removes the ambiguity perfectly. I repudiate most 
emphatically the statement attributed to me by Mr. ALLIs 
(’03, p. 662) that in any fishes nerve hillocks (neuromasts) 
may be innervated by communis fibers. |The exact opposite I 
regard as one of the most important and distinctive of the re- 
sults which I have reached in my studies upon the nerve com- 
ponents of fishes. 

A.tts further argues for the similarity of the two systems 
of sense organs on the ground of the resemblance in the mode 
of the distribution of ampullary organs in very young elas- 
mobranchs to that of terminal buds in adult ganoids (Amia) 
and teleosts. This resemblance in distribution pattern un- 
doubtedly exists, and AtLtis’ demonstration of the origin of 
ampullary organs in the ontogeny of Mustelus diffusely scattered 
over the cutaneous surface in the positions of the mouths of their 
pores in the adult is a point of no small interest and impor- 
tance. His description of pit lines in Mustelus in relations simi- 
lar to the pit lines of ganoids and teleosts is also of importance, 
showing, as I fully agree, that the ampullary organs cannot be 
homologized with these pit lines. 

But it by no means follows because the ampullae are not 
homologous with the pit lines that they are therefore homolo- 
gous with the terminal buds. In fact, my studies of Ameiurus 
have shown that in this type, in addition to lateral line canals 
and pit lines of strictly typical arrangement, there are two in- 
dependent systems of diffusely scattered cutaneous sense organs, 
which I have termed the small pit organs and terminal budsand 
which are innervated by lateralis and communis nerves re- 


Herrick, Zerminal Buds of Fishes. — 135 


spectively. If we must seek for equivalents of selachian ampul- 
lary organs, they may be found in these small pit organs, for the 
two sets of organs agree in all essential points of structure and 
innervation, save that in the adult the teleostean organs do not 
sink down into deep tubes and become massed at their inner 
ends into dense clusters, but retain more nearly the embryonic 
condition of the selachian organs. A careful review of all of 
the known facts will show that the two types of sense organs 
(terminal buds and lateral line organs of various types) 
may appear or vanish quite independently of each other. 

In Autis’ last paper (02, p. 663) he rebuts JOHNSTON’s 
criticisms in the following language. I have numbered the 


items in the passage quoted for ease of reference. 

*¢ (1) That end-buds are all innervated by fibers that ‘find their 
central endings in the lobus vagi;’ (2) that all other forms of cutaneous 
sense organs are innervated by fibers that ‘have their central ending in 
the nucleus funiculi, tuberculum acusticum, or the cerebellum;’ (3) 
that the respective centers for the lateral line and end-bud fibers are so 
separate and stable ‘that it is utterly impossible for fibers or centers to 
‘undergo modification” of any sort such as I understand ALLIs to 
mean ;’ (4) that ‘It is impossible that these organs [end-buds and lat- 
eral line or pit organs] should ever resemble one another in any other 
than a superficial way ;’ (5) and that end-buds are organs ‘with visceral 
function (e.g. taste),’ while all other sense organs are organs ‘with a 
somatic function (e.g. touch, &c.)’, are certainly nothing more nor less 
than deductions from the theory he seeks to establish in his several 
works instead of established facts on which to base that theory.” 

Now, nothing could be further from my desire than to en- 
ter into this controversy in a spirit of captious criticism, yet I 
think it can clearly be shown that there is a solid foundation in 
fact for nearly every one of the contentions taken by JOHNSTON 
in the passage quoted. And it should be clearly born in 
mind that many of the most decisive facts to which I refer have 
been brought to light since Mr. AtLtis first formulated his 
theory of the genetic relationship between ampullary organs 
and terminal buds. : 

Now taking up the abuve points in serial order: (1) 
We have accurate observations on this point in Acipenser 
(JoHNSsToN, ’O1), Petromyzon (JouNsTon, 02), Gadus (HEr- 


RICK, 00) and Ameiurus (HerRRICK, OI, the lobus faciali 


136 JOURNAL OF COMPARATIVE NEUROLOGY. 


ing a derivative of the lobus vagi) and I also have unpublished 
observations on Anguilla, Carassius and other fishes in sufficient 
numbers to make it plain that it is true generally that in all 
teleosts possessing terminal buds these organs are innervated 
from the lobus vagi or its derivative, the lobus facialis. And 
there are no precise observations on the other side. 

(2) The second point I think is abundantly established by 
the works of all recent students of nerve components, particu- 
larly those of StrRonG, JoHNSTON and HeErRIck. 

(3) The third point is stated perhaps rather more strongly 
than the facts at command permit; nevertheless, I think it 
must be admitted that the metamorphosis of organs of touch, 
particularly tactile hairs, into end organs for the perception of 
mass movements of water (lateral line organs; cf. PARKER, ’03), 
for the maintenance of bodily equilibrium (semi-circular ‘canals) 
and for hearing presents far less of difficulty than the transfor- 
mation of any of the organs of this series into gustatory organs. 
For the organs of the first series are all stimulated by impacts 
of a common type, differing only in mode of application, 
rythmic character, etc., while the gustatory organ belongs to a 
totally different modality. 

(4) The fourth point we have touched upon above. What- 
ever may be conceived abstractly as ‘‘possible,”’ it is clear that 
in point of fact these two types of organs do not resemble each 
other either in structure, innervation or central connections. 

(5) Finally that the terminal buds are organs of taste is no 
longer a matter of conjecture, but a fact of proof. 

We may, then, summarize our examination as follows: 
The morphological rank and functional significance of the terminal 
bud system of sense organs is definitely fixed. They are in no 
way related to any organs of the lateral line system (pit 
organs, nerve hillocks, ampullary organs, etc.) but on the other 
hand they are most intimately related to the taste buds within 
the mouth. This relationship is shown by their identity in 
structure, innervation, central connections and functions. On 
the other hand, the phylogeny of the terminal buds is by no 
means on so secure a foundation as that of the sense organs of 


HerRIcK, Zermznal Buds of Fishes. 137 


the acustico-lateralis system and much remains here for future 
research to clear up. Wecan however advance with confi- 
dence at least this negative conclusion, that the terminal buds 
have not been derived from ampullary organs or any other 
members of the acustico-lateralis system. 


Denison University, 
May 7, 1903. 


MIT ERATURE. Cl LED: 


ALLIS, EDWARD FHELPS. 
or. The Lateral Sensory Canals, the Eye-Muscles, and the Peripheral 
Distribution of certain of the Cranial Nerves of Mustelus laevis. 

OS Via Soles ND 

03. On Certain Features of the Lateral Canals and Cranial Bones of 
Polyodon folium. Zool. Jahrb. Abt. f. Anat., XVII, 4. 

BATESON, W. 
*90. The Sense Organs and Perceptions of Fishes, with Remarks on 
the Supply of Bait. Journ. Biol. Assoc., London, Vol. 1. 
CLAPP, CORNELIA, M. 
’99. The Lateral Line System of Batrachustau. /ourn. Morph., XV, 2. 
HERRICK, C. JUDSON. 

’99-~=0 The Cranial and First Spinal Nerves of Menidia; A Contribu- 
tion upon the Nerve Components of the Bony Fishes. Journ. Comp. 
Neurol., 1X, 3-4. 

700. A Contribution upon the Cranial Nerves of the Cod Fish. 
Journ. Comp. Neurol., X, 3. 

’o1r. The Cranial Nerves and Cutaneous Sense Organs of the North 
American Siluroid Fishes. Journ. Comp. Neurol., XI, 3. 

702. The Sense of Taste in Fishes. Abstract. Sczence, N. S., XVI, 
No. 400. 

Houser, G. L. 

’o1. The Neurones and Supporting Elements of the Brain of a Se- 

lachian. Journ. Comp. Neurol, XI, 2. 


JounstTon, J. B. 
?o1. The Brain of Acipenser. Zool. Jahrb., Abt. f. Anat., XV. 
’o2. The Brain of Petromyzon. /ourn. Comp. Neurol., XII, 1. 
‘o2 a. An Attempt to Define the Primitive Functional Divisions of 
the Central Nervous System. /ourn. Comp. Neurol., XII, I. 
702 b. The Homology of the Selachian Ampullae. A Note on Allis’ 
Recent Paper on Mustelus laevis. Azat. Anz., XXI. 
KInGSpury, B. F. 
97. The Structure and Morphology of the Oblongata in Fishes. 
Journ. Comp. Neurol., V\1. 1. 


138 JOURNAL OF CoMPARATIVE NEUROLOGY. 


LEYDIG, FR. 

51. Ueber die fussere Haut einiger Siisswasserfische. Zetts. f. wiss. 
Zool., 11. 

’79. Neue Beitraige zur anatomischen Kenntniss der Hautdecke und 
Hautsinnesorgane der Fische. Festschr. Naturf. Ges. zu Halle. 

’94. Integument und Hautsinnesorgane der Knochenfische. Weitere 
Beitrige. Zool. Jahrb. Abt. f. Anat., VIII, 1. 

MERKEL, FR. 

*So. Ueber die Endigungen der Sensiblen Nerven in der Haut der 

Wirbelthiere. Rosvock. 
MorriL., A. D. 

95. The Pectoral Appendages of Prionotus ant their Innervation. 

Journ. Morph., X\. 
PARKER, G. H. 

’03. The Sense ot Hearing in Fishes. Am. Nat., XXXVII. 

’03. Hearing and Allied Senses in Fishes. WU. S. Fish Commyssion 
Bulletin for 1902, Washington, 1903. 

SCHULZE, F. E. 

763. Ueber die becherférmigen Organe der Fische. Zezts. f. wiss. 
Zool., X11. 

’67. Epithel- und Driisenzellen. Arch. f. mikr. Anat. II. (Argues 
that terminal buds on the outer skin of fishes are gustatory and serve to 
localize the source of the stimulus.) 

’70. Ueber die Sinnesorgane der Seitenlinie bei Fischen und Amphi- 
bien. Arch. f. mtkr. Anat., VI. 

STRONG, O. S. 

‘03. The Cranial Nerves of Squalus acanthias. Abstract. Sczence, N. 

So x VILS No 424. 


ON TE NAUK (OF. THE PERICELLULAR NET- 
WORK OF INERV EE (CELAS: 


By SHINKISHI HATAI. 


(from the Neurological Laboratory of the University of Chicago.) 


With Plate III. 


In 1895 HeELp announced that the nerve cells in the cen- 
tral system are densely surrounded by the terminals of the 
axones which divide into fine branches and form by a local 
union a complicated network. He also described the club- 
shaped enlargement of the axone terminals which may be seen 
in the embryonic nervous system. These terminals, or ‘‘Axen- 
cylinderendflache,’’ are characterized by the presence of the 
minute granules or neurosomes which stain red by erythrosin. 

The terminal network or ‘‘Pericellular network,’”’ or ‘‘Gotuer- 
netz of BETHE” is very well shown by BeruHe’s molybdenum 
technique, as well as by Gorai’s modified silver technique. 
The appearances produced by the. foregoing methods are 
somewhat different from those obtained by the method of 
Hep. This difference in appearance may be due to the fact 
that both Gouer’s and Brrue’s technique cover over the finer 
and more complicated structures by the precipitation of silver 
chromate or the molybdenum compound respectively, and thus 
bring out strongly marked and rather angular meshes of the 
network. The nature of these networks obtained by HE Lp, 
Gore! and Berue will "be described later on. 

I have also noticed and described this pericellular network 
which not only surrounds the PuRKINJE cells and the cells in 
the corpus trapezoideum, but also the cells in the Ammon’s 
horn and those in the ventral cornua of the spinal cord (03). 

HeELp’s idea that the finer network which surrounds the 


140 JoURNAL OF COMPARATIVE NEUROLOGY. 


cell-body and :orms the pericellular network, is composed of 
the terminals of the axones, and that the granules which stain 
red are seuroplasm, has been criticized by ApATHy. APATHY 
thinks the Axencylinderendflache to be nothing more than 
the neuroglia fibers which very frequently surround the 
cell-body as well as the dendrites, and thus denies the nervous 
nature of HeELp’s pericellular network. 

He says: ‘‘In Betreff der von HELD beschriebenen peri- 
cellularen Ausbreitungen des Axencylinder und in Betreff der 
Neurosomen desselben Autors, kann ich indessen versichern, 
dass erstere nichts mit dem Axencylinder, letztere nichts spezi- 
elleres mit dem Nervosen tberhaupt zu thun haben. Jene 
Ausbreitung sind ein Gliagitter (die Neuroglia im urspriing- 
lichen weiteren Sinne verstanden), welches von dem Axen- 
cylinder, den es wahrend seines Weges im Centralnervensystem, 
ausserhalb der Myelin Scheide, umhiult, auf den Zellkorper der 
Ganglienzelle tibergeht und sich auf die sonstigen Auslaufer der 
Ganglienzelle, auf die Dendriten, fortsetzt. Ein ahnliches, die 
Ganglienzellen eng umschliessendes Gliageflecht habe ich auch 
bei Hirudineen beschrieben und es als die Gliazone der Gangli- 
enzellen bezeichnet, welche auch in das innere der Ganglienzelle 
Fortsatze senden kann, aber nicht eigentlich zur Ganglienzellen 
gehort und nicht mit dem Neurofibrillengitter zu verwechseln 
ist. Durch dieselbe Gliazone der Wirbeltierganglienzelle ist, 
wie ich glaube, auch die von Gore! durch Chromsilber darge- 
stellte und urlangst beschriebene Gitterhille bedingt.”’ 

This criticism of ApATHy is, no doubt, one objection to 
HE Lp’s hypothesis, since, as has been pointed by ApArny, the 
cell-bodies are intimately surrounded by the neuroglia fibers. 
Furthermore, HELp’s technique is not suited to distinguish the 
neuroglia fibers from that of the pericellular network, although 
he claims that these two structures stain differently. 

Recently HeEtp published an elaborate article ('03) in which 
he not only criticised the results obtained by various investiga- 
tors, but also confirms fully his own earlier observations. He 
employed several different methods for both staining and fixing, 
especially Gotci’s modified silver method and Berue’s tech- 


Harta, Nature of the Pericellular Network. 141 


nique were used. The paper is accompanied by a large num- 
ber of drawings made from his own preparations. In some 
cases, he noticed a direct continuation of the pericellular net- 
work by way of terminals with the medullated nerve fibers. 
Thus he proves conclusively that one set of structures forming 
the pericellular network is composed of the terminals of the 
axones, and, in other words, this particular network is nervous 
in nature. BETHE (00) also showed a direct continuation of 
his ‘‘primitive neurofibrils’’ into the Golginetz. BETHE here 
used his own molybdenum technique. It is therefore evident 
that the weight of opinion is in favor of the nervous nature of 
the network and against the view of APATHY. 

Another objection to the view of HELp has been raised by 
BETHE (00). Although he agrees with HeELp in considering 
the pericellular network to be composed of the terminals of the 
axones represented by the primitive neurofibrils, yet he does 
not accept the delicate meshwork structure formed by the 
union of the side branches of the axone terminals as well as 
the presence of the neurosomes. HELp’s answer to this criti- 
cism is not a satisfactory one and demands further evidence. — 
My own results obtained from the studies on the preparations 
stained with a new technique, reveal several facts which support 
very strongly HeEtp’s idea as well as my own observations pre- 
viously published ('03), and are opposed to the views of 
ApATuy and BETHE. 

The staining reagents used for the present investigation 
were prepared by the following way : 

Minced sheep’s brains were boiled in 10% formalin for 
several days on the water-bath. The solution was filtered 
off while hot. A concentrated aqueous solution of ammonium 
molybdenum in excess was added together with a few drops of 
HCl and the solution boiled again. The solution is at first 
yellow and then turns to a blue color. After continued boiling 
(at least 24 hours) an intense blue color is produced. The na- 
ture of the substance which gives the blue color has not yet 
been determined. The preparation is stained with the blue so- 
lution thus obtained, in the following way: The materials are 


142 JouRNAL OF COMPARATIVE NEUROLOGY. 


fixed with 10% formalin and the sections are made by either 
paraffin or celloidin technique according to the usual procedure. 
The sections are first treated for several minutes with dilute hy- 
drochloric acid, and then transferred to the blue solution where 
they remain for 24 hours. If the section is over-stained, it 
_may be decolorized with 1% potassium permanganate for a 
second, and the further oxidation is stopped by transferring the 
section into dilute hydrobromic acid. The preparation thus 
stained brings out the structures described below. 

Unless otherwise mentioned, the following description is 
based on the observations of cells from the spinal cord of the 
dog. <A careful study of these preparations shows that the 
nerve-cell-body, as well as dendrites, are stained intense blue, 
while the axone appears stained alight blue. The neurokeratin 
in the medullary sheaths, as well as neuroglia fibers stain bluish 
black; the former being more intensely stained. The blood 
vessels and nuclei of their walls, as well as the neuroglia nuclei 
stain faintly blue, and may easily be distinguished from the 
others. The terminals of the axones stain more faintly than 
the axones near their origin from the cell-body, although some 
of the contents of the nerve terminals or neurosomes stain an 
intense blue. Thus my technique brings out the several struc- 
tures at once in different tones, and enables us to analyze the 
material forming the grey matter. 

As Fig. 1 shows, there are a large number of the medul- 
lated nerves which were cut in different planes. In all cases we 
can distinguish these fibers from their surroundings by utilizing 
the fact that the outer layer of the medullated sheath contains 
well marked neurokeratin which stains an intense black. A 
large numberof the dendritic branches are also found inter- 
mingled with the medullated nerve fibers just mentioned. 
These dendritic branches are also cut in various planes and are 
distinguished from the axones by the fact that they are desti- 
tute of the medullated sheaths and at the same time they are 
densely surrounded by the neurosomes. If we examine the 
cross-section of a dendrite, a large number of the granules 
which surround its circumference are easily observed. Another 


Hatal, Nature of the Pertcellular Network. 143 


structural element to be mentioned here, is the neuroglia fiber. 
These fibers are easily found with my stain, since it stains them 
an intense black, and also on account of their filamentous char- 
acter. They are distributed thickly around the nerve fibers, 
dendritic branches, blood vessels, etc., but their well-marked 
characteristics enable them to be readily recognized. 

Besides these structures just mentioned, together with 
blood vessels and neuroglia nuclei, there are left in the grey 
substance very important structures, the axone terminals or 
fine filaments which carry the neurosomes. These terminal fil- 
aments are extremely difficult to isolate from the surrounding 
tissue, since they stain very faintly. To meet this difficulty, it 
is necessary to examine first the neighborhood of the cell- 
bodies where, as a rule, a lymph space is formed surrounding 
the latter, and where also the axone endings alone occur. As 
we shall see from Fig. 1, a large multipolar cell is surrounded 
by the clear space where a large number of the terminals (Fig. 
I, 2) carrying the neurosomes are seen. A single bundle of 
the axone terminals in this drawing comprises a number of fine 
filaments which unite and run together. From this locality, if 
we trace the terminals toward the grey substance, the faintly 
stained bluish filaments containing neurosomes are distinguished. 
These are the terminals which we are trying to demonstrate. 

I have pointed out already in my previous paper (’03), 
that the motor cells in the spinal cord are densely surrounded 
by the terminals of the axones which form the so-called _peri- 
cellular network. The network just mentioned is well shown 
in Fig. 3. These terminals, as Figs. 1 and 3 show, come 
in contact with the cell-body where they form very com- 
plicated meshes. These meshes not only cover the whole sur- 
face of the cell-body, but also extend continuously along the 
entire course of the dendrites. As I have said already, in the 
substantia grisea, the cross and longitudinal sections of the 
dendrites cut in various planes and having different sizes, are to 
be observed. Some of the sections are so minute that they 
can be seen only with the highest powers. Even such spicules 
of the dendritic branches are surrounded by the meshes of the 


rs 


144 JOURNAL OF COMPARATIVE NEUROLOGY. 


axone terminals. This proves that the dendrites are entirely 
covered by the meshes of that network. The axone as well as 
medullated fibers are differently characterized, since they are 
entirely free from the coverings of these terminals. This fact 
is well shown in Fig. 2. The relation of these meshes to 
- either dendrites or the cell-body is shown by both Figs. 1 and 
2. Fig. 3 represents one process of the dendrites of the motor 
cells of the cat, stained with HEIDENHAIN’s iron-haematoxylin. 
As we see from these two figures, the filaments form an ex- 
panded meshwork containing a large number of neurosomes 
and come in contact with the cell-surface. These meshes unite 
again with those of the neighborhood and completely cover the 
cell-body and dendrites. A part of this meshwork on the cell- 
surface is shown in Fig. 3, which has been drawn from a motor 
cell of a cat stained with HerpeNnnain’s iron-haematoxylin. 
The axone terminals, in some cases, form a coarse network in 
the neighborhood of the cell-body (Fig. 1), or in some in- 
stances, these meshes are formed after the axone terminals have 
reached the surface of the cell-body (Fig. 2). 

Thus my preparation shows that the pericellular network 
is composed exclusively of the substance of a nervous nature 
and there is not the slightest evidence that the glia fibers 
take part in the formation of the meshes. This confirms the 
observations previously published. This network which I am 
treating of coincides with that obtained by HE Lp (’95) and 
AUERBACH (99). The question now arises whether this net- 
work will coincide also with that obtained by Gotar (’98) and 
BETHE (’00). 

By means of an elaborate technique, BETHE was able to 
show a peculiar reticular structure covering the cell-bodies as 
well as dendrites. To this reticular structure, just mentioned, 
BETHE (’00), gave the name ‘‘Golginetz 
the ‘‘Golginetz”’ is homologous with structures described by 
Meyer, HEtp and AverBacH. He says: (1) ‘‘Alle Gang- 
lienzellen und ihre Protoplasma Fortsatze (auch in den feinsten 


”) 


According to him, 


Verzweigungen) sind umgeben von specifischen Netzen, welche 
gung g 
ich ‘‘Golginetze’”’ nenne. Eine Ausnahme bilden hierin die Spinal- 


Hatal, Nature of the Pericellular Network. 145 


ganglienzellen, die Zellen der aufsteigenden Trigeminuswurzel 
und des Lobus electricus des Zitterrochens, bei denen bisher 
derartige Netze nicht nachgewiesen werden konnten. Auf die 
Axencylinder (Neuriten) geht das Golginetz nicht tiber. (2) Die 
Golginetze sich berithrender Neurone sind meist einander durch 
Netzmaschen verbunden. (3) Bisweilen kann man Axencylin- 
derzweige direkt in die Golginetze tibergehen sehen.” 

Thus BETrHe was able to seea direct continuation of the 
axis-cylinder to the Golginetz. HELD was also able to see such 
a direct continuation of the two structures just mentioned. 
BETHE, however, believes that the Golginetz is composed of 
the combined bundles of the ‘‘Primitive Neurofibrils’’ which 
form the meshwork by interweaving with one another. He 
strongly opposes the view of HELp who regards the meshwork 
as composed of fine branched filaments which carry the neuro- 
somes. BETHE further denies the existence of the neurosomes 
which, according to him, are produced by the action of the re- 
agents employed. HeELp answers BETHE’s assertion in his 
most recent article (00). There HELD says that there are two 
structures in the pericellular network; one is composed of an 
entirely non-nervous substance or neurokeratin, while the sec- 
ond is the one which is composed of the nervous substance 
or axone terminals. The neurosomes are contained within the 
second nervous network only. He.further says that BreTHE’s 
Golginetz coincides very probably with the non-nervous net- 
work. As a consequence of the technique which he employed, 
BrETHE was unable to see neurosome granules as well as the fine 
meshwork. According to Hetp the non-nervous network is 
derived from the neurokeratin or ‘‘Gliaschnurringe’”’ which ap- 
pear surrounding the medullated sheath at certain intervals. 
He p illustrates the direct continuation of the Gliaschnirringe 
with the neurokeratin meshwork. Concerning the second criti- 
cism offered by BeErTuHeE, that is, the statement that the neuro- 
somes might have been formed as artifacts through the action 
of the chemicals employed he says that there is no reason to 
think that the neurosomes are an artificial product, since all the 
general histological methods for nerve cells bring out this structure 


146 JOURNAL OF COMPARATIVE NEUROLOGY. 


in the same size, form, appearance, and distribution. I also have 
given some evidence in an earlier paper, that the neurosomes can 
hardly be regarded as an artificial product, and so far as modern 
histological technique is concerned, there is no reason to favor 
BETHE’S view on this point. 

However, I cannot agree with HELD who thinks that there 
are two kinds of network, one nervous and the other non- 
nervous, and that the former may coincide with BETHE’s Golgi- 
netz. HELp’s non-nervous network has been demonstrated by 
using the molybdenum technique of BrErHr. According to 
HELD, this technique brings out so-called ‘‘Gliaschnirringe ” 
which send filaments to form the non-nervous network. This 
point seems to me quite doubtful, for if the Gliaschnuirringe as 
well as the non-nervous or neurokeratin network were stained, 
one would expect to find neurokeratin structures along the 
medullary sheath. The latter, however, were not observed by 
him. My own technique, which is able to stain both neuro- 
keratin and glia, fails to show the existence of the non-nervous 
or neurokeratin network about the cell-body. One can see fre- 
quently the neuroglia fibers which surround the cell-body ; 
these, however, merely pass over the cell-surface and never stop 
on it. I noticed very often the neurokeratin rings or ‘‘Gliasch- 
nirringe’ of the medullary sheath in my preparations but was 
unable to observe any process from these rings such as has 
been described by Hetp. Therefore I am unable to confirm 
HELD on this last point. 

From the above I believe that there is only one kind of 
pericellular network which is formed by the terminals of the 
axones, and further, I believe that GoLGi's or BETHE’S network 
is identical with the network described by HELD as the nervous 
network from the (1) coincidence or indentical distribution, and 
(2) from the reticular structure observed in all cases. A differ- 
ent appearance shown by different technique, may be due to 
the fact that the Gore! and BerHe methods produce a precipt- 
tation of silver chromate and molybdenum salts respectively, 
which covers the finer meshwork, thus obscuring the internal 
structure and giving it a coarse appearance. 


Hatal, Nature of the Pericellular Network. 147 


BIBLIOGRAPHY. 


798. APATHY: Bemerkungen zu Garbowski’s Darstellung meiner Lehre 
von den leitenden Nervenelementen. Azologzesches Centrat- 
blatt, 1898, Vol. XVIII, pp. 703-713. 


’98. GOLGI: Sur Ja structure des Cellules nerveuses. Arch. ttal. de 
Biol., Turin, t. XXX. 


99. AUERBACH: Das terminale Nervennetz in seinen Beziehungen zu den 
Ganglienzellen der Centralorgane. pp. 190-214. Monat- 
schrift f. Psychiatrze. Bd. V1. 


’00. BETHE: Ueber die Neurofibrillen in den Ganglienzellen von Wir- 
belthieren und ihre Beziehungen zu den Golginetzen. 
Archiv f. mikros. Anat. Bd. 55. pp. 5!3-544- 


’02. HELD: Ueber den Bau der grauen und weissen Substanz. drch. 
Anat. u. Entwickl. Jarg. 1902. Heft 5-6. pp. 189-224. 


03. HaTAl: The Finer Structure of the Neurones in the Nervous 
System of the White Rat. Decennzal Publications, University 
of Chicago. 1903, Vol. X. 


ILLUSTRATIONS ON PLATE III. 


fig. ¢.—Ventral horn cell of a dog. 
Fig. 2.—Ventral horn cell of cat. Iron-haematoxylin. 
Fig. 3.—Dendritic branch of ventral horn cell of cat. Iron-haematoxylin. 


a = neuroglia. 

b = blood-vessel. 

c = medullated nerve fiber. 
d = axone terminals. 

e = neuroglia nuclei. 

f — cell body. 

g = dendrites. 


* 


THE JOURNAL OF COMPARATIVE NEUROLOGY, VoL. XIII. BEATE IIIS 


ca 


; Y . ( Ne ee Y / peat 


ay S 


EX 
Be 
| 
< oaiae, . O he ee ~ 
yo NW g GEE FD 
SOOT Ge 
A Woe ee Aes 
® by re / 
cox 


ti NEUROKERATIN IN’ THE MEDULLARY 
Site brs (OR THE) PERIPHERAL 
NERVES OF MAMMALS. 


By SHINKISHI HATAI. 


(From the Neurological Laboratory of the University of Chicago.) 
With Plate IV. 


At as early a date as 1876, EwaLp and KUHNE announced 
that the nervous system contains a horn-like substance which 
has a great resistance to chemical reagents and especially to 
gastric and tryptic digestive fluids. Onaccount of its similarity 
to the horny material found in other tissues, they termed the 
substance obtained from the nervous system ‘‘Neurokeratin.”’ 
JOSEPHINE CHEVALIER, in 1884, demonstrated the presence of 
the neurokeratin in the medullary sheath of the peripheral 
nerve fibers, by means of chemical analysis. Later (1890) 
CHITTENDEN again took up the problem of the neurokeratin with 
Professor KUuNEg, the first investigator of it, and not only con- 
firmed the previous statement of EwaLp and KUuHNE, but at the 
same time cleared up many obscure points. In addition to the 
chemical determination of it, they showed also that the neu- 
rokeratin network in the medullary sheath may be demonstrated 
under the microscope by treating the fiber with boiling alco- 
hol, ether and digestive fluids. By examining under the micro- 
scope the slides thus prepared KUHNE and CHITTENDEN arrived 
at the conclusion that there are two layers of the neurokeratin 
network in the medullary sheath; one closely surrounding the 
axis cylinder, the other lying immediately beneath the primi- 
tive sheath. 

Through the investigation of these writers just mentioned, 
the new field of the study was opened for the neurologist. 
There are a large number of articles concerning the neuroker- 


I50 JOURNAL OF COMPARATIVE NEUROLOGY. 


atin, but it is not necessary to review them in detail. Generally 
speaking, two different opinions are held by the recent investi- 
gators; one, according to which the neurokeratin is not a pecu- 
liar substance, but an artefact produced by the reagents em- 
ployed, an albuminous substance (RANVIER) or a nucleoproteid 
(Wynn). This view is held by several histologists, for instance, 
ENGELMANN, GERLACH, KOLLIKER and LAvpowsky. The sec- 
ond view, that the neurokeratin is a peculiar substance in the 
nervous system and preexists as such, is held by most of the 
physiological chemists as well as by a number of histologists ; 
for instance, Lrypic, PALUDINO, JOSEPH, SCHIEFFERDECKER, 
KapLan and others. 

In recent years the use of WEIGERT’s and HEIDENHAIN’S 
haematoxylin has given many opportunities to study the struc- 
ture of the neurokeratin, and as a consequence, a large number 
of reports have already been made. Most of the recent inves- 
tigators are inclined to believe in the preexistence of the neu- 
rokeratin in the medullary sheath, on both chemical and histo- 
logical grounds, and moreover the complexity of the neuroker- 
atin increases as the technique improves. 

Before discussing the literature further, I shall give an ac- 
count of my own results. 

For this investigation, the nerve fibers were preserved in 
10% formalin and cut into sections 1oyv thick after the usual 
paraffine technique. The sections were stained with the ‘‘blue 
solution’”’’ and decolorized sufficiently with diluted NHsOH. 
In some cases the blue solution was used without subsequent 
decolorization. 

The preparations thus treated show the double layers of 
the neurokeratin network in the medullary sheath, one layer 
closely surrounding the axis cylinder, while the other lies just 
beneath the primitive sheath as was stated by EwaLp and 
KUxNE and others. Such an arrangement of the neurokeratin 


1 Preparation of the ‘‘blue solution’’ and its application to the microscopical 
sections is given in my other paper,—‘‘ On the Nature of the Pericellular Net- 
work.”? Journ. Comp. Neurol., Vol. XIII, No. 2, p. 141. 


Harat, Sheaths of the Peripheral Nerves. 151 


can best be seen by examining the cross-section of the nerve 
fibers. 

Fig. 1 shows the cross-section of the peripheral nerve of 
the cat, treated by the method mentioned above. The neuro- 
keratin stains an intense pink, while the rest of the structure 
remains for the most part unstained ; the axis cylinder presents, 
however, a faint blue tint. It may easily be seen that each 
fiber shows the double rings of the neurokeratin, one along the 
outer margin and the other along the inner margin of the me- 
dullary sheath. These rings, however, are not complete cir- 
cles, but appear as broken lines or are composed of dots. The 
size as well as number of the dots in each ring is variable. In 
some cases, instead of minute dots, a continuous long line is 
noticed, and in still other cases filaments more or less tangential 
to the inner ring are observed. The latter appearance is more 
common in the rings which immediately surround the axis 
cylinder. In some cases, these two rings are connected with 
each other by means of threads of neurokeratin which run per- 
pendicularly to the long axis of the fibers and in cross-sections 
give an appearance of radiations from the central ring; in other 
words, wheel-like structures are formed by means of connecting 
bands between the two rings. The manner of the distribution 
of the connecting band is so irregular that it is almost impossi - 
ble to give a general description of all the different appearances. 
I can merely state that the number of the connecting bands just 
mentioned is not constant, sometimes six or seven being ob- 
served and in some cases more than a dozen. Furthermore, 
the spaces formed by the connecting bands are also variable in 
size and shape. Some of the connecting bands are united with 
others by means of the delicate side-branches which they give 
off. In some instances, three rings are noticed (Fig. 1,4). The 
reason for this irregular distribution of the neurokeratin in the 
cross-section of the nerve fibers is easily understood if one ex- 
amines the longitudinal sections of the same fibers, and to these 
we now turn. 

The longitudinal view of the nerve fibers is shown in 
Figures 2, 3 and 4. The preparations were made from the peri- 


152 JOURNAL OF COMPARATIVE NEUROLOGY. 


pheral nerves of the dog, using the same technique as was em- 
ployed for the cross-section of the fibers. 

Fig. 2 shows the general arrangement of the nerve fibers, 
as well as the manner of distribution of the neurokeratin. In 
this figure, the well known cone-shaped or funnel-shaped struc- 
tures are shown clearly. The cone, as is known already, is 
formed by the network of the neurokeratin; the basal or wide 
end attaches to the outer layer of the neurokeratin sheath, 
while the narrow tip is attached to the neurokeratin sheath 
which directly surrounds the axis cylinder. In many cases, in- 
stead of the single cone there is the appearance of two cones 
united by their apices and at their bases attached to the outer 
margin of the sheath, as is shown in Fig. 2. The shape, size, 
direction, as well as number in each internode of the nerve 
fiber, is not constant. In some fibers the distance between the 
two narrow tips of the cones is quite regular, and the median 
longitudinal section of such fiber shows it to be divided into 
equal segments by means of the neurokeratin funnels or cones. 
(Fig. 2, 7). Both the outer and inner layers of neurokeratin 
network are continuous throughout the entire length of the 
nerve fiber, as they are not interrupted at the nodes of Ran- 
VIER (Fig. 2, a,g.) The structures of the cones and their rela- 
tion to the two sheathing layers are shown in Fig. 4. I have 
mentioned already that double rings of the neurokeratin are no- 
ticed in examining the cross-sections of the fibers, and further 
gave the hint that the dots forming the rings are connected by 
means of delicate filaments of the same substance, and these 
connecting filaments are in turn continuous by means of the 
side-branches. Such a complicated arrangement may be under- 
stood by examining the structures drawn in Fig. 2 and Fig. 4. 
Since the two sheathing layers of the neurokeratin are con- 
nected either by radial filaments from the inner sheath toward 
the outer layer or by the branches which are sent off from the 
inner surface of the cones toward the inner neurokeratin sheath, 
there is thus formed a complicated arrangement of the neuro- 
keratin in the cross-sections of the fibers. In most cases, the 
outermost layer of the neurokeratin has fine meshes, as is 


Hara, Sheaths of the Peripheral Nerves. 153 


shown in Fig. 2, 6, while in some cases larger spaces of the 
network are noticed at the small end of the funnel (Fig. 3) 
while the basal portion of the funnel has much finer meshes. 
Wyyn ('00), who studied the medullary sheath by using 
the PaL-WEIGERT technique, states that each cone is composed 
of six segments, placed at regular distances apart, and converg- 
ing from the primitive sheath to the axis cylinder. The pres- 
ent writer has examined a large number of preparations 
stained with Pat-WEIGERT technique, but fails to confirm this 
result, although, in some cases, there are six dots placed at 
regular distances (Fig. 1,a) and located in some cases along the 
inner margin of the primitive sheath and sometimes closely sur- 
rounding the axis cylinder. A careful examination, however, 
reveals that in most cases more than six dots are to be seen, 
and furthermore, each dot is not a single dot but composed of 
several irregular lines and minute dots, thus giving a structure 
such as is to be seen in my preparations. According to 
Wiassak’s (00) investigation, the PaL-WEIGERT technique 
stains only the substance known as cerebrin, which is a con- 
stituent of the myelin, but not of the neurokeratin. If this 
statement of WLassak is true, the Pat-WEIGERT technique 
shows a distribution of the cerebrin in the medullary sheath, 
but not the distribution of the neurokeratin. The six rods de- 
scribed by Wynn, therefore, may prove that the cerebrin in the 
medullary sheath is distributed in a special relation to the 
cones ; in other words, the cerebrin and neurokeratin are closely 
interdependent in their distribution. Wyww arrived, however, 
at the conclusion that the substance which forms the cones is 
neither a fatty body, nor neurokeratin, but a nucleoproteid. 
The statement given by Wynn, that the six dots given in the 
medullary sheath brought out by Pat-WeEIGErT technique, are 
not the neurokeratin, seems very probable, for by comparing 
sections by the Pat-WEIGERT method with mine for neuro- 
keratin, the dots of the former are seen to be of a size compara- 
ble with the spaces formed by the neurokeratin net in my slides 
and thus the two preparations give complementary pictures 
(Fig. 5). Passing by the contradictory statements of Wynn 


154 JOURNAL OF COMPARATIVE NEUROLOGY. 


and WLAassAK, it is only necessary for me to accept the state- 
ments that the cones described by WyNN are not true neuro- 
keratin but a substance (either neucleoproteid or cerebrin) pre- 
cipitated along the walls of funnels, and as a consequence, to 
conclude that Wynn is dealing with some substance other than 
neurokeratin. As a consequence, Wynn failed to obtain the 
network structure of the neurokeratin. The results obtained 
by the Pat-WEIGERT technique are, however, extremely inter- 
esting, for the preparations stained by this method reveal the 
manner of distribution of the cerebrin (WLassaxk) which is com- 
plementary to that of the neurokeratin and thus indicate in- 
directly the distribution of the neurokeratin. 

The structure of the medullary sheath of the peripheral 
nerves, as revealed by these sections, seems to be as follows: 

The neurokeratin network which contains the myelin con- 
sists, first, of two thin layers, one beneath the primitive sheath, 
the other around the axis cylinder; these two layers are, how- 
ever, continuous at the nodes of RANVIER; second, of a chain 
of cones connected with both layers, the base of a cone being 
attached to the peripheral layer, the apex in the inner layer, 
each cone being made up of a continuous band of neurokeratin 
exhibiting meshes of variable size and converging from the 
primitive sheath to the axis cylinder. The clefts of Scumiprt- 
LANTERMANN occur at the situation of the cones, but in some 
cases they are also visible between the two cones.  Further- 
more, the cones present a great variability in number, size, and 
direction. These phenomena, just mentioned, may indicate 
that the Scumrpt-LANTERMANN clefts are produced artificially by 
rupture of the coagulated protoplasm along the walls of the 
cones. My own observations described above may advan- 
tageously be summarized as follows: 

(1) The peripheral nerve-sheath contains two layers of 
the neurokeratin, one beneath the primitive sheath and the 
other along the axis cylinder. These two layers are connected 
by bands of neurokeratin which run obliquely from the outer to 
the inner layer, thus presenting a funnel or a cone-shaped ap- 
pearance. 


Hara, Sheaths of the Peripheral Nerves. 155 


(2) The neurokeratin sheath has a large number of pores 
or meshes and presents a net-like appearance. The size and 
‘shape of the meshes are, however, highly variable. 

(3) Neither the outer nor inner layers of the neurokeratin 
are interrupted at the node of RANvirrR, but become continuous 
through it with the corresponding layers of the adjoining inter- 
nodes. 

(4) The present observations do not agree with Wynn's, 
but confirm the previous statements of EwaLp and KUHNE. 

(5) The present technique brings out a more detailed 
structure of the neurokeratin than has been obtained by other 


investigators. 


BIBLIOGRAPHY. 


EWALD and KUHNE: Ueber einen neuen Bestandtheil des Nervensystem. 
Verhand. d. naturh. med. Vereins zu Heidelberg. Vol. 1, Heft 5, 1876. 


CHEVALIER, JOSEPHINE: Chemische Untersuchung der Nervensubstanz. 
Zettsch. f. Physiol. Chem., Bd. X, 1884. ; 


KUHANE and CHITTENDEN: On Neurokeratin. Zhe Mew York Medtcal 
Journal, Feb. 22 and March 1, also in Zezt. f. Bzol., Bd. 26, 1890. 


KOLLIKER: Handbuch der Gewebelehre. Bad. II, 1896. 


WLASSAK: Herkunftder Myelin. Arch. 7. Entwicklungsmechanthk der Or- 
ganismen. Vol. VI, Heft IV, 1808. 


Wywn, W. H.: The minute structure of the medullary sheath of nerve- 
fibers. Journ. Anat. Phystol., Vol. 34, 1900. 


KAPLAN, L.: Nervenfairbungen (Neurokeratin, Markscheide, Axencylin- 
der). Arch. f. Psych. u. Nervenhetdlkunde, Bd. 35, Heft 3, 1900. 


The above bibliography is supplementary to the references given by K6L- 
LIKER, WYNN and KAPLAN. 


156 JoURNAL OF COMPARATIVE NEUROLOGY. 


EXPLANATION OF THE FIGURES. 
PIGAWE Sve 
The following figures are free-hand drawings, using Zeiss Oc. 4 x obj. 1-12. 


Fig. 1.—Cross-sections of the ventral root nerves of the cat, showing dis- 
tribution of neurokeratin. 


‘ig. 2.—Longitudinal section of the ventral root nerves of the dog, show- 
ing form of cones. 


Fig. 3.—Largely magnified drawing of a single nerve-fiber taken from the 
preparation used for Fig. 2, showing the meshwork in the cones. 

Fig. g.—Same as Fig. 3. 

Fig. 5.—Enlarged view of a cross-section of a single nerve-fiber taken from 


the preparation used for Fig. 1, showing neurokeratin framework. 


fig. 6. Cross-section of ventral root nerve of human spinal cord. PAL- 
WEIGERT. Largely magnified. Showing masses of medullary substance in- 
cluded in neurokeratin framework. 


PLATE IV. 


JOURNAL OF COMPARATIVE NEUROLOGY. VoL. XIII. 


iy 
yy’ 

Ag died # 

Fig.5 
or. 
c 

Fig. 6 


Fig. 2 


ta 

i 

is £ 
Fig. J 


! 


Ch] 


an Wig fey Osh, : 

SW ae VIR Peekecy £32 Se gt 
" eee > 34° ay” eid at 
Ce Ot OB} ot Bee O 


s 

re 

é 
1@ 


\e 
Fig.3 


LITERARY NOTICES. 


Hearing and Allied Senses in Fishes.! 


The moot question whether fishes possess the sense of hearing has 
been attacked experimentally by PARKER with the result that he gives 
a definite affirmative answer for some types of teleosts. This report 
will impress the reader as a particularly clean piece of experimental re- 
search and the conclusions seem to be free from any reasonable ques- 
tion. ‘The experiments were ingeniously planned, carefully controlled 
and skilfully performed. The general methods of the research, as well 
as the conclusions, can be gathered from the author’s summary of re- 
sults, which we quote: 

1. Normal Fundulus heteroclitus reacts to the sound waves from a 
tuning-fork of 128 vibrations per second by movements of the pectoral 
fins and by an increase in the respiratory rate. It probably also re- 
sponds to sound waves by caudal fin movements and by general loco- 
motor movements. 

2. Individuals in which the eighth (auditory) nerves have been 
cut do not respond to sound waves from the tuning-fork. 

3. The absence of responses to sound waves in individuals with 
severed eighth nerves is not due to the shock of the operation or to 
other secondary causes, but to the loss of the ear as a sense organ. 

4. Fundulus heteroclitus therefore possesses the sense of hearing. 

5. ‘The ears in this species are also organs of prime importance 
in equilibration. 

6. Normal Fundulus heterochtus swims downward from the top of 
the water and remains near the bottom when the aquarium in which it 
is contained is given a slight noiseless motion. 

7. Individuals in which the nerves to the lateral-line organs have 
been cut will swim upward or remain at the top while the aquarium is 
being gently and noisclessly moved. 

8. The lateral-line organs in this species are probably stimulated 


1 Parker, G. H. Hearing and Allied Senses in Fishes, U. S. Fish Com- 
mission Bulletin for 1902, Washington, 1903, pp. 45-64- 


xxii JouRNAL OF ComPaARATIVE NEUROLOGY. 


by a slight mass movement of the water against them. They are not 
stimulated by sound waves such as stimulate the ears. ‘ 

g. Individuals in which the nerves to the lateral-line organs have 
been cut swim downward and thus escape from regions of surface wave 
action. ‘They also orient perfectly in swimming against a current. 
Since surface waves and current action stimulate fishes in which the 
nerves to the lateral-line organs and to the ears have been cut, these 
motions must stimulate the general cutaneous nerves (touch). 

10. The vibrations from a bass-viol string when transmitted to 
water stimulate the ears and the lateral-line organs of Fundulus. They 
also stimulate mackerel and menhaden, but not the smooth dog-fish, 
which responds only when in contact with solid portions of an aquarium 
subjected to vibrations. 

We have another interesting set of observations in Professor 
TULLBERG’S paper on the Functions of the Labyrinth in Fishes.’ 

The author operated on various teleosts by cutting the semi-circu- 
lar canals, removing the large otolith and in various other ways and 
concludes that the labyrinth is not an organ of equilibrium, or of the 
static sense or for the maintenance of muscular tone or of the 
spacial sense in the sense of v. Cyon. It is probably to some degree 
an auditory organ (though he gives no satisfactory proof of this). 
‘Originally and primitively, however, the labyrinth of fishes is a sense 
organ for the perception of movements of the surrounding water, since 
currents are apparently perceived by the cristae acusticae of the am- 
pullae, but wave movements probably by the maculae acusticae of the 
utriculus, the sacculus and the lagena. The central organ for this 
sense organ is apparently the cerebellum.” 

That the labyrinth is primarily an organ for the perception of cur- 
rents or streaming movements of the surrounding medium seems on 
a priort grounds highly improbable, since the stimulus is one which 
may easily act upon the skin, lateral-line organs or other superficial 
sense organs, but only with difficulty on the deep-seated labyrinth. 
Moreover the experiments cited seem inconclusive. In the first 
place, the canals were merely cut and the nerve endings were not de- 
stroyed. The same applies to the removal of the otolith. In several 
of these experiments there were forced movements and disturbances of 
equilibrium which the author has to explain away. In general too the 
lesions were symmetrical. More radical operations would seem to be 


? TycHo TULLBERG. Das Labyrinth der Fische, Ein Organ zur Empfindung 
der Wasserbewegungen. SBithang till K. Svenska Vet.-Akad. Handlingar, Stock- 
holm, Bd. 28, Afd. IV, No. 15, 1903. 


Literary Notices. XXill 


necessary before permitting the conclusion that the labyrinth is not 
concerned in equilibration and the static sense. In experiment 12, 
designed to show that orientation with reference to water currents is 
not done by the lateral-line canals, the n. lateralis vagi was cut behind 
the shoulder girdle, with the result that the operated fishes still oriented 
themselves with reference to currents like normal fishes. But it should 
be noted that in this experiment the canals of the head, of far greater 
extent and importance, were uninjured. 

Professor TULLBERG’S experiments are criticised at some length by 
Dr. PARKER in connection with a brief report upon his own experi- 
ments recently published in the American Naturalist. ! 

A physiological and morphological classification of all of the cuta- 
neous sense organs of fishes as conceived by the present writer is now 
in press in the current number of the .\merican Naturalist, and the 
status of such of these organs as belong to the communis or gustatory 
system is treated more at length in another place in this issue of this 
Journal. Fe Be 


Taste Fibers and Their Independence of the Trigeminus.? 


The surgical work and clinical observations upon which this re- 
port is based seem to have been more carefully planned and more skil- 
fully wrought out for the solution of the problem of the course of the 
taste fibers than any of the preceding contributions to this difficult 
theme. In most of these cases a preliminary test of gustatory sensi- 
bility was made before the operation—a most necessary precaution, as 
the event proved. ‘The patients were whenever possible kept under 
observation and repeatedly tested tor long intervals after the operation. 
The results in all of the cases furnish a strikingly clear proof of the the- 
sis stated in the title, without the confusion and ambiguity of most pre- 
vious reports. 

In general there is a post-operative transient period of total or 
partial abolition of taste perception, with a gradual return to the nor- 
mal gustatory sensibility, but no return of tactile sensibility. He says, 
“T find it difficult to reconcile my fairly uniform results, that is, uni- 
form in so far as the ultimate preservation of taste is concerned, with 
the contradictory observations which have been made by so many 


* PARKER, G.H. The Sense of Hearing in Fishes. Am. Nat., XXXVII, 
No. 435, March, 1903. 

* CusHING, Harvey. The Taste Fibers and their Independence of the 
N, Trigeminus. Deductions from Thirteen Cases of Gasserian Ganglion Extir 
pation. Johns Hopkins Hospital Bull., X1V, No. 144, 145, 1903, pp. 71-78. 


XXiVv JouxNnaL oF ComPARATIVE NEUROLOGY. 


others lhe only explanation which I can offer is that there may be 
in a considera -ercentage of the cases a temporary diminution in 
its acuity or a complete abolition of taste as has been intimated above, 


and that this may have been interpreted under certain circumstances 
as an evidence of permanent loss of this sense.” 

The conclusions are: 1. That the perception of taste is unaf- 
fected on the posterior portion of the tongue and never permanently or 
completely lost on its anterior two-thirds after removal of the Gasserian 
ganglion. 

2. ‘That the temporary abolition or lessening of the acuity of 
taste may be found to exist over the anterior and anaesthetic portion of 
the tongue for some days after the operation. 

3. That this temporary loss of function may possibly be occa- 
sioned by some interference with chorda transmission brought about 
by a mechanical or toxic disturbance due to degeneration of the N. 
lingualis. 

4. Thata lesion of the trigeminal nerve may be associated with 
disturbance of taste over the chorda territory without the necessary in- 
ference that the nerve is a path for gustatory impulses. 

5. That the N. trigeminus in all probability does not convey 
taste fibers to the brain either from the anterior or posterior portion of 
the tongue. . 

This last conclusion, it will be noted, has been reached almost 
uniformly of late by researchers in three independent lines of work, 
the morphologists, the embryologists and the comparative anatomists, 
and it is a source of satisfaction to see the confusing clinical evidence 
at last brought into harmony with these in so unambiguous a manner. 

Cee 
De Fursac’s Psychiatry.’ 
Almost every language has had its little compend based on the re- 
markable changes in psychiatric views, produced by rather a daring 
but decidedly inspiring reform started by KRAEPELIN. This little book 
adopts the classification of KRAEPELIN, which is partly etiological, but 
largely a grouping of the mental diseases according to their outcome. 
The little book has in some respects an intrinsic value, owing to the 
attempts at harmonizing current French views with those of the Heidel- 
berg school. The book is dedicated to Professor Jorrroy, and is in a 
way a semi-official acknowledgement of KRAEPELIN’S attitude. 
A. M. 


1 Manuel de Psychiatrie, by Dr. RoGues De Fursac. farts, F. Alcan, 
1403. 


Literary Nottces. XXV 


The Evolution of Man and his Mind.! 


This bulky volume, which contains much that is admirable, is de 
signed to be a popular exposition of the course of human evolution, 
particularly from the point of view of certain sociological defects of 
our present status. This makes a striking background for an exposé 
and arraignment of certain corrupt tendencies in our political and so- 
cial organization. ‘The literary style is colloquial and catchy and the 
book should do good in directing popular attention toward these abuses. 
As a whole, however, it is so ill-balanced and full of inaccuracies that 


it can hardly be commended as a helpful scientific contribution. 
CL Me 


The Brain and Nerves of the Anamnia.? 

Professor JOHNSTON’s summary of recent progress in our knowl- 
edge of the central and peripheral nervous system of the Ichthyopsida 
is one of the notable papers of the year. The point of view from 
which he writes is so distinctively his own that his article is more than a 
mere abstract or critical review ; it is a positive contribution toward the 
solution of some of the major problems of comparative morphology. 
This point of view he has already presented in this Journal (March, 
1902), under the title, ‘An Attempt to Define the Primitive Functional 
Divisions of the Central Nervous System,” being essentially the corre- 
lation of central with peripheral differentiation of the nervous system. 
We venture the prediction that the next decade will see this principle 
worked out successfully in several fields at present open merely to the 
methods of descriptive anatomy. ‘lhe conclusions of students of 
nerve components as formulated at the present moment may or may 
not stand the test of time, but the esseutial aims and methods of work 
which they have introduced into comparative neurology will surely in 
the end yield results of permanent value. Professor JOHNSTON's 


Referat is therefore very timely. Gi Fi 


1 CLEVENGER, S. V. The Evolution of Man and his Mind. A History 
and Discussion of the Evolution and Relation of the Mind and Body of Man 


and Animals. Chicago, 1903. 

2 Jounsron, J. B. Das Gehirn und die Cranialnerven der Anamnia. 
Merkel und Bonnet’s Ergebnisse, Bd. XI, for wor.  MWresbaden, 1902, pp. 973- 
1852. 


XXVi JouRNAL OF COMPARATIVE NEUROLOGY. 


The Dorval Spino-cerebellar Tract.' 


Sections of the sprmal cords of dogs at different thoracic and cervi- 
cal levels show that the longest fibers of the’ direct dorso-lateral cere- 
bellar tract, i. e., those arising in the lowest levels of the spinal cord, 
are most superficial in position and that the shorter fibers are added 
successively along the inner side of this zone. 

The authors verify previous findings of degeneration in the cells 
of CLARKE’s column below the lesion after section of the dorso-lateral 
cerebellar tract. In these cases the fibers of the tract between the de- 
generated cells and the lesion show no degeneration under the MARcul 
procedure, WEIGERT stain, anilin blue-black, picrocarmine, etc. To 
test the condition of these fibers further the authors made a right lat- 
eral transection of the Xth thoracic segment 260 days subsequent to a 
total transection at the IId thoracic level. The animal (dog) was sac- 
rificed 20 days after the establishment of the second lesion. The 
right cerebellar tract above the second lesion was found fully present- 
ing all the signs of WALLERIAN degeneration under the MARCHI re- 
action. The left cerebellar tract appeared normal and without any 
degeneration. ; 

‘It would seem therefore that atrophy, severe and long-lasting, 
probably permanent, of CLAaRKE’s cell-column induced by spinal tran- 
section in the the lower cervical or upper thoracic region, far from de- 
stroying the dorsal cerebellar tract, leads to no obvious or easily de- 
monstrable degeneration of at least the main body of the fibers of the 
tract. Further, after the severe atrophy of CLARKE’s column has set 
in and become established, transection of the tract in the very region 
of atrophy of the cell-column still causes full WALLERIAN degeneration 


of the fibers head-ward of the transection.” These results obviously 
have an important bearing on the theory of the physiology of the neu- 
rone and its biological interpretation. CC. 4 


The Optic Chiasma in Symmetrical and Asymmetrical Teleosts.* 


The author in continuing his observations on the optic chiasma of 
fishes develops several results which bear directly onthe morphology 
and phylogeny of the flat fishes. He has previously shown’ that in 


* SHERRINGTON, C. S.and LAstetr E. E. Remarks on the Dorsal Spino- 
cerebellar Tract. Journ. of Physiol., XXIX, 2, March, 1903. 


2 Parker, G. H. The Optic Chiasma in Teleosts and its Bearing on the 
Asymmetry of the Heterosomata. Aull. Mus.;Comp. Zool., XL, 5, 1903. 


’ The Crossing of the Optic Nerve in Teleosts. Biol. Bull., 11, 1901, pp. 
335-336. 


Literary Notices. XXVii 


the symmetrical teleosts the right and left optic nerves are dorsal in an 
approximately equal number of cases. He now extends the observa- 
tions to the flat-fishes and finds that the two families into which the 
sub-order Heterosomata is divided by recent systematists differ con- 
spicuously in this character. In the Soleidae the chiasmata are di- 
morphic, as in symmetrical teleosts; i. e., the right optic nerve is dor- 
sal in abeut half of the observed cases and ventral in about half. In 
the Pleuronectidae, on the other hand, the chiasmata are monomorphic 
for each species; in dextral species the left nerve is dorsal, in sinistral 
species the right nerve is dorsal. All species of this family that turn 
in only one direction have their dorsal nerves connected with their 
migrating eyes. In all species that have both dextral and sinistral in- 
dividuals, the dorsal nerve is connected with that eye which in the 
greatest number or in the nearest of kin migrates. The unmetamor- 
phosed young of the Pleuronectidae are not symmetrical in the same 
sense that symmetrical teleosts are, for they have monomorphic chias- 
mata. ‘The Soleidae are not degraded Pleuronectidae, but degenerate 
descendants of primitive flat-flshes, from which the Pleuronectidae 
have probably been derived. The monomorphic condition of the 
optic chiasma of the Pleuronectidae can be explained only on the as- 
sumption of natural selection. The flat-fishes afford striking examples 
of discontinuous variation. ae Fear” 


Brain Weights of Eminent Men.’ 


Dr. SpirzKa has tabulated 96 cases and made comparisons with 
the statistics of 800 brains of ordinary persons as given by BISCHOFF 
and MaRCHAND. ‘‘ The average (arithmetical) brain weight of the 96 
individuals is 1473 grams, exceeding the various averages given for the 
European brain by 75 to 125 grams, and this without allowing for the 
advanced age of this series; the average of 92 being 63 years.” ‘‘Itis 
further shown that the period of decrease [in brain weight] with age is 
deferred for fully a decade among the more intellectual persons, a 
point already alluded to by Dona.pson, and significant in connection 
with the longevity of healthy persons endowed with high intelligence.” 
The paper is accompanied by several instructive curves and tables. 

5 ee 


* Spitzka, Epwarp ANTHony. A Study of the Brain Weights of Men 
Notable in the Professions, Arts and Sciences. Philadelphia Medical Journal, 
May 2, 1903. : * ot th 


XXVili Jounnat or Comparative NeuRoLocy. 


The Lateral Neasory Syvtem of the Eels.' 


in this paper of 48 pages and three good"plates we have a detailed 
description of the lateral line canals and associated sense organs in the 
conger ee] and a more brief account of three related species of Mu- 
raenidae, viz., Ophicthys serpehs, Myrus vulgaris and Muraena helena. 
The nerves supplying these sense organs have been traced only in their 
peripheral portions and we are promised a later research upe the i in- 
nervation of these structures in the conger. 

In connection with the innervation of the pit organs shete is is one 
correction of minor importance on page 42 of the reprint, to which at- 
tention might be drawn. Mr. ALLis describes a pit line running paral- 
lel with the squamosal lateral canal which is innervated by a nerve 
formed from an anastomosis between a branch of the facial, which is re- 
garded as a portion of the ramus opercularis facialis, and a branch of 
the glossopharyngeus or vagus. In discussing the morphology of these 
pit organs the author says, ‘‘If they be pit organs, it is practicully cer- 
tain that they cannot be innervated by the facialis, for there is no sin- 
gle instance that I know of, of lateral sensory fibers accompanying the 
ramus opercularis of that nerve.” As amatter of fact I have described 
just such a condition in Menidia (this Journal, vol. 1X, p. 294, seqq.), 
BauDELoT’s descriptions strongly suggest the same thing, and I have 
no doubt that a careful search of the literature would reveal other such 
cases. Confusion arises in this connection (and this is the occasion of 
this note) from the fact that there are two opercular rami in teleosts 
which are not always distinguished. ‘The ramus opercularis profundus 
VII is a purely motor nerve, and this is the only one of these branches 
which is mentioned by Srannius. On the other hand, the ramus 
opercularis superficialis VII is a mixed nerve, containing in Menidia 
both general cutaneous and lateralis fibers and in other cases appa- 
rently it may contain communis fibers also or be fused with the motor 
fibers belonging to the ramus hyoideus VII. This superficial nerve 
frequently anasomoses with the vagus or glossopharyngeus and in 
Menidia at any rate it is clear that the vagal ae are all of general 
cutaneous nature. C.J. Be 


1 Auuis, E, P. The Lateral Sensory System inthe Muraenidae. J#éern. 
Monatsschrift f. Anat. u. Phystol., XX, 4-6, 1903. 


VotuME XIII. 1903. NUMBER 3. 


THE 


JournaAL oF Comparative Neurovoey. 


THE NEUROFIBRILLAR STUCTURES IN THE GAN- 
GEA’ OR THE LEECH (AND: CRAYFISH WITH 
ESPECIAL REFERENCE TO THE NEURONE 
RHEORY. 


By C. W. PRENTIss, 
Parker Fellow in Zoéblogy, Harvard University. 
With Plates V and VI. 


A previous publication treats in detail of the fibrillar net- 
works in the neuropil of the leech. In the present paper it is 
purposed to give amore general description of the fibrillar struc- 
tures found in the nervous system of the leech and crayfish and 
to point out the relation of these structures. to the neurone 
théory. 

The neurone theory, grounded upon the fundamental re- 
searches of GuDDEN, GotGI, and His, was first formulated by 
WALDEYER (’g!) in the following words: ‘‘Das Nervensystem 
besteht aus zahlreichen, unter einander anatomisch wie gene- 
tisch nicht zusammenhangenden Nerveneinheiten (Neuronen).”’ 
As WALDEYER, VERWORN, and Nisst have shown, the all-impor- 
tant point embraced in the neurone theory is not the anatomical 
independence of the nervous elements, but the assumption that 
the nervous system is entirely composed of cell individuals. 
Whether the processes of these cells are only in contact, or by 
growing together have become continuous, is a secondary mat- 
ter. Nevertheless, on the threefold evidence of histogenesis, 
neuropathology, and histology, most neurologists maintain that 
the nervous system is composed of anatomically independent, 
cellular units. 

1. Hustogenesis. That the nerve elements develop from 
single neuroblasts and not from chains of cells was first asserted 


158 JoURNAL oF CoMPARATIVE NEUROLOGY. 


by His (’89, ’90). His work has recently been confirmed by 
the excellent research of Harrison (:01). The latter was 
able to show that in two cases dorsal processes from bipolar 
neuroblasts of the sensory ganglion broke through the bound- 
ing membrane of the nerve chord, while ventral processes from 
the same cells were traced to the periphery. Both His and 
Harrison have proved that the axis-cylinders and dendrites of 
motor elements originate as processes from a single cell; they 
traced the axis-cylinder processes to the very point where the 
nerve fibers appear; but in no case did either investigator de- 
monstrate direct connection between these processes and the 
embryonic nerve fibers. Harrison states moreover that: ‘‘Die 
ersten motorischen Fasern sind schon vorhanden, ehe wuber- 
haupt lose Zellen in der Gegend der Austrittsstelle zu finden 
sind.” 

Directly contrary to the observations of His and Harri- 
SON, and in agreement with those of BALFour (’76) and DoHRN 
(91) are the preliminary statements of BeTHE (: 02) as to the 
histogenesis of the nerve elements in the chick: (1) Before the 
axis-cylinder processes of the neuroblasts break through the 
bounding membrane of the chord, the fundaments of nerve 
fibers are formed as chains of cells; (2) coincident with the 
breaking through of the processes, many primitive fibers may 
be observed in the myotomes; (3) processes of the bipolar 
cells which form these nerve fibers in the myotomes may be 
traced into the chord with the same distinctness with which the 
processes of the neuroblasts (of His) may be traced out of it, 
and often the union of processes from neuroblast and primary 
nerve cell may be observed; (4) the primitive nerve fibers are 
differentiated simultaneously from an entensive chain of cells 
extending from the central organ to the periphery; (5) these 
cells increase in number only by karyokinesis ; not until the 7th 
to goth day of development are the neuro-fibrillae formed. 
BETHE concludes from these observations that each nerve ele- 
ment represents a group or society of cells, rather than a single 
cell individual. His statements are as yet unsupported by pub- 
lished figures, but they agree both with the observations of 


PRENTISS, Weurofibrillar Structures. 159 


other noted neurologists and with his own recent work on the 
regeneration of peripheral nerves. It is at least clear that as 
far as the evidence of histogenesis goes the neurone theory is 
still open to dispute. It is grounded on certain recognized facts, 
but these facts relate only to the early stages of development. 
Neither His nor Harrison says anything as to the origin of the 
neurofibrillae, structures upon which the opponents of the reu- 
rone theory put much weight. 

2. Neuropathology. GUDDEN (’89) was the first to demon- 
strate the fact that the cutting of a motor axis-cylinder caused 
the degeneration not only of the peripheral fiber, thus isolated 
from its cell, but also of the cell and its dendrites. In new-born 
animals the entire nerve element atrophied and was resorbed, 
but never in any case were the pathological changes observed 
beyond the dendrites of the injured neurone. According to 
Niss_, Foret (’86) first coupled these facts with the evidence 
of GoLGi’s preparations and formulated the idea of the nerve- 
cell individual, to which WALDEYER later gave the name of neu- 
rone. The two facts of neuropathology which have been used 
as arguments in support of the neurone theory are: (1) that 
nerve fibers separated from their ganglion cells degenerate and 
(2) that the phenomena of degeneration never have been ob- 
served to pass beyond the processes of the injured elements. 

The experiments of GupDEN show that it is not merely 
their isolation from their cells which causes nerve fibers to degen- 
erate, for the cells themselves often atrophy in young animals. 
BETHE ('98) by isolating the neuropil of a nerve center in the 
brain of the crab found that the nerve elements may remain 
actively functional for several days, proving that the nerve ele- 
ments are physiologically independent of their cells. His re- 
cent experiments in neuropathology, of which he has given 
only a preliminary account (: 02), show that peripheral nerves 
will degenerate some time after isolation from their nerve cells; 
furthermore, that in the young dog such degenerate nerve fibers 
will, in the course of 6 to g months, regenerate all the structures 
of anormal nerve fiber—primitive fibrillae, perifibrillar substance and 
Schwann's sheath. Not only are the regenerated fibers normal 


160 JOURNAL OF COMPARATIVE NEUROLOGY. 


in structure, but stimulation of the distal stump of the nerve 
(Ischiadicus) causes contraction of the muscles which it sup- 
plies. Upon dividing the regenerated nerve a second time, 
degeneration ensued in the distal portion only. By this evi- 
dence of BETHE’s the phenomena of degeneration, in themselves, 
are rendered worthless as arguments in support of the neurone 
theory, while his observation of the regeneration of peripheral 
nerves is incompatible with the assumption that each nerve 
fiber develops in its entirety as a process of a single ganglion 
cell. 

That the phenomena of degeneration have not been ob- 
served to pass beyond the dendrites of the injured nerve ele- 
ments is most easily accounted for by assuming the non-exist- 
ence of continuity between the dendrites of the nervous ele- 
ments. But, as Nisst argues, the opponents of the neurone 
theory may, with equal right, assume the presence of connecting 
fibrillar structures in the central organ, of whose peculiar qual- 
ities we as yet know nothing, at least in vertebrates. These 
structures (the ‘‘nervose Grau” of Nissx) are differentiated cell 
products, and as they are independent of the ganglion cells, 
they are immune from the pathological changes which affect the 
processes of the latter. What are the histological facts in sup- 
port of these assumptions ? 

3.  Astology. We have seen that the neurone theory 
was based by Foret and WaLDEYER mainly upon the discov- 
eries of His, and the silver impregnations of GoLai and Ramon 
y CajaL. It is now known that the methods of GOLGI are ex- 
tremely unreliable, that the impregnations are rarely complete 
and often extend to non-nervous structures. When, therefore, 
ApATuHy ('97) demonstrated by new methods the finer structure 
of the nervous system, many neurologists maintained that 
the neurone theory had received its death blow. The 
value of ApATHY’s work, however, has been in throwing into 
doubt the evidence of GoLG! preparations, and showing the 
importance of more certain and refined methods for the study 
of the nerve elements. APATHY proved that the supposed 
nervous units of GoLGi preparations are themselves composed 


PRENTISS, Weurofibrillar Structures. 161 


of infinitely smaller conducting elements, the primitive fibrillae, 
which form networks in the cells. He also figured cases of 
direct communication between the processes of nerve cells in 
the intestine of Pontobdella and instances of fibrillar networks 
in the neuropil of A/zrudo. 

In addition APATHY maintains the existence of large motor, 
and small sensory, fibrillae; these two types of fibrillae are 
connected with each other by networks in the ganglion cells 
and by the diffuse fibrillar network which, according to APATHY, 
forms the neuropil proper. 

The existence of the neurofibrillae is now generally admit- 
ted. BETHE (’98) confirmed ApAtruy’s observations as to the 
presence of the fibrillae in both the nerve cells and fibers, but 
could not distinguish between motor and sensory fibrils in the 
crab, Carcinus. He also asserts that the fibrillar networks in the 
neuropil are not diffuse. His series of studies on the neuro- 
fibrillae in the nervous elements of vertebrates (’98a, '99, : 00) 
leads him to the conclusion that they are invariably present, 
but that the networks characteristic of invertebrate nerve cells 
are rarely found. He suggests that the neurofibrillae of the 
cells may be directly connected with the fibrillar ‘‘Golginetze’’ 
which surround the cells, and that these in turn may be in com- 
munication with the fibrillae of other (sensory) elements. 
Neither BETHE nor Nissi was able to demonstrate a clear case 
of such fibrillar connection, and the assumption that the 
‘‘Golginetze’”’ are composed of neurofibrillae has been severely 
criticized by Ramon y Caja and others. NIssL (: 03) assumes 
the existence, in the central organ, of ‘‘nervous gray”’ struc- 
tures, the differentiated products of nerve cells, corresponding 
perhaps to ApATHY’s diffuse fibrillar network. The assumption 
that such nervous elements exist, is based entirely on theoreti- 
cal grounds. He points out that it has never been proved that 
the neurofibrillae of the nervous system are integral parts of 
the nerve cells, but that there are facts which indicate that they 
are not: (1) the fibrillae are sharply marked off from the rest 
of the cell in both structure and staining qualities; (2) the axis- 
cylinders are prolonged far beyond the limits of the cell proto- 


162 JOURNAL OF COMPARATIVE NEUROLOGY. 


plasm; (3) in diseased and degenerate cells certain fibrillar 
tracts, which pass in and out through the dendrites, may remain 
intact and cannot be distinguished from normal fibrillae. Nuissv’s 
figures are schematic and, as in the text, it is difficult to sepa- 
rate fact from theory. His book, however, is of great value 
in that it discloses the weak spots in the neurone theory and 
shows that little or nothing is known of the extra-cellular ele- 
ments found in the gray substance of the vertebrate nervous 
system. Between the dendrites of the nerve cells and the 
point at which the sensory axis cylinders lose their medullary 
sheaths, there is practically a total blank in our present knowl- 
edge of the nervous elements. 

Nissi accepts as an established fact ApATuy’s hypothesis 
that the nerve elements of invertebrates are connected by a 
diffuse network in the neuropil. Not so the supporters of 
the neurone theory, who are, however, divided in opinion. By 
far the majority of them admit the existence of the neuro- 
fibrillae, but deny that there is continuity between the neurones. 
Prominent among this school are von LENHOSSEK (’g9Q), S. 
MEYER ('99), and VAN GEHUCHTEN (:00). Other neurologists, 
like WaLDEYER, HocuHE (’99) and VERWORN (:00), while ad- 
mitting that fibrillar continuity may exist, hold, and we think 
rightly, that the question of contact or continuity between the 
neurones is a side issue. They doubt the existence of fibrillar 
‘‘Gitterwerke,’’ however, and still maintain that the nervous 
system is composed entirely of cell units. 

Because of this doubt which still exists in the minds of 
many neurologists, as to certain of ApATHy’s observations, the 
writer has made a special study of the fibrillar structures found 
in the neuropil of H/zxudo, the results of which are now in press. 
The present paper furnishes further evidence as to the structure 
of the neuropil and is supplemented by a more genera! study 
of the neurofibrillae in the nerve elements of both A#udo and 
Astacus. The research was begun at the Zoological Laboratory 
in the University of Freiburg, Baden, and was completed at the 
Strassburg Physiological Institute. 


PRENTIsS, Neurofibrillar Structures. 163 


Material and Methods. 


The ventral ganglia of the leech (Hrudo medicinalis) and 
the abdominal ganglia of the crayfish (Astacus fluviatilis) formed 
the material on which most of my study was based. A part of 
the Hzrudo material was treated as described in my former 
paper (PRENTISS, : 03), the method being based on that of 
BETHE (: 00a). Sections 10u thick, fixed in corrosive sublimate 
were impregnated with ammonium molybdate solution (1:4000- 
1:6000), differentiated about one minute in warm water (55°— 
60° C) and then stained with an aqueous solution of toluidin 
blue (1:3000). Inthe ganglion cells a pure fibrillar stain was 
obtained by fixing ganglia for one hour in ether fumes, staining 
zm toto with toluidin blue (1:3000) and fixing the stain ina 1% 
solution of ammonium molybdate. The material was then 
dehydrated, embedded in paraffin and sectioned in the usual 
manner. This method is simple, but uncertain in its results. It 
is a selective method, like methylen blue, and not all of the 
fibrillae are demonstrated. Often, however, preparations were 
obtained which showed the fibrillae with diagrammatic dis- 
tinctness. 

The preparations of Asz/acus material were all stained zntra 
vitam with methylen blue. The fibrillae were differentiated by 
leaving the ganglia 2-4 hours in normal salt solution ; the stain 
was then fixed in ammonium picrate, which differentiates the 
fibrillae more clearly than molybdate. 


The Fibrillae in the Ganglhon Cells. 


ApATHY describes two types of cells in the ventral ganglia 
of Hirudo, distinguished from each other by their size and the 
structure of the neurofibrillae. In the type to which the 
smaller cells belong, one large fibril enters the cell and forms a 
close meshwork of rather large fibrillae about the nucleus. 
This is a motor or cellulifugal fibril according to APATHY; its 
network about the nucleus is connected by radial fibrils with a 
finer peripheral meshwork formed by smaller cellulipetal or 
sensory fibrillae. In this manner he assumes that sensory and 
motor elements are put into direct communication within the 


164 JOURNAL OF COMPARATIVE NEUROLOGY. 


cell itself. The cells of the second type are the largest in the 
ganglion. Their fibrillae are of nearly equal size and forma 
diffuse network throughout the plasma of the cell. 

The difference in the fibrillar structures contained in these 
two types of cells was very apparent in my own preparations. 
A good example of the smaller type (the motor cells of 
ApAtuy) is illustrated in Figure 14 (Plate VI). The inner 
network about the nucleus was quite distinct in the preparation; 
the fibrillae of which it is composed are somewhat larger and 
therefore easier to trace than those in the periphery of the cell, 
but preparations in which practically all of the fibrillae within 
the cells are stained, show no such sharp distinction in the size 
of the fibrillae as ApAtTuHy describes. Certain fibrils might 
appear a little larger than others, but my preparations do not 
warrant the assertion that the larger fibrillae a/ways form the 
inner network, and the smaller the outer one, as APATHY main- 
tains. Such large fibrillae as APATHy describes are often found, 
however, in preparations in which only a portion of the fibrils 
are stained. It may also be observed that the smaller the 
number of the fibrillae to be seen in a cell process, the larger 
those fibrillae usually are. It is well known that APATHY’s 
gold chloride method demonstrates the fibrillae more com- 
pletely in the cells than in their processes, and the large 
‘‘motor” fibril which ApAtuy figures entering the cell and 
forming the inner network, is, in every case, it may be noted, 
the only fibril in the cell process. Several large fibrillae are often 
formed in the cell processes by the cleaving together of the 
primitive fibrils; if the impregnation of these were incomplete 
so that only one fibril is visible, conditions would be produced 
like those figured by ApATHy. We have such a case evidently 
in Figure 3 (Plate V). One large fibril (4) is seen in the cell 
process; this, however, divides into four smaller ones on enter- 
ing the cell. It is obvious that but few of the fibrillae are 
stained, which renders it especially easy to follow those which 
are demonstrated. This cell belongs to the smaller type 
described by ApAruy, but it may be observed that only one 
branch (a) of the large fibril joins the inner network about the 


PRENTISS, eurofibrillar Structures. 165 


nucleus; the others take a more peripheral course, a fact 
which is not in agreement with ApATuHy’s figures. In the same 
section another cell of the same type was found (Figure 2, 
Plate V). In this case, however, there are many fibrils of 
nearly equal size in the cell process, although the networks 
within the cell are incompietely stained. Neither in verte- 
brates nor in crustacea do the neurofibrillae of the nerve cells 
show any marked correlation in size and function. 

In the giant cells of the leech there is no inner network 
about the nucleus; the fibrillae are very numerous in the cell 
process and divide to form a network of small irregular meshes 
throughout the greater portion of the cell. A portion of such 
a cell is shown in Figure 4 (Plate V) and gives some idea of the 
great number of fibrillae which these giant cells contain. 

The fibrillar structures in the ganglion cells of Astacus 
and other decapod crustacea resemble somewhat those of the 
larger type of cells in Azrudo. Such structures have been 
described by BEeETHE (’98), and OwsIANNIKOW (: 00). BETHE 
describes the fibrillae within the ganglion cells of Carcinus as of 
nearly equal size, and forming a network of somewhat large 
meshes throughout the plasma of the cell. OwsIANNIKOW 
finds primitive fibrillae of two sizes in the nerve cells of 
Astacus ; the smaller of these are found about the nucleus in 
the form of a network; the larger fibrils occupy the peripheral 
portion of the cell, and are the continuations of the fibrillae in 
the cell process. 

My preparations of Astacus showed no trace of a network 
of fine fibrillae about the nucleus. The usual condition ob- 
served is seen in Figure 15 (Plate VI). The fibrillae appear 
relatively large and form a few large meshes in the peripheral 
region of the cell. This figure corresponds to the descriptions 
of BerHEe and resembles the only drawing which Owsranni- 
KOW gives of the neurofibrillae in the cells; the figures of the 
latter do not support the statements he makes in the text. 


Fibrillar Structures tn the Cell Processes. 


BETHE (’98) was the first to observe that in the crab a 


166 JOURNAL OF COMPARATIVE NEUROLOGY. 


nerve element may contain a greater number of neurofibrillae 
than are found entering its cell, and to show that this was due 
to the fact that many neurofibrillae may enter a neurone 
through the collaterals, and pass out either through the 
peripheral fiber or through other collaterals without entering 
the cell proper. He regards such conditions as incompatible 
with the neurone theory, because the fibrils which do not enter 
the cell can not be integral parts of it, but must be the product 
of some other cell or cells. The fibrillar structures in the 
nerve elements of both the leech and the crayfish confirm 
BETHE’S observations. From methylen blue preparations of 
the abdominal ganglion of Aséacus, certain important facts may 
be observed without the use of a high magnification. As 
ApaATHY noted in the leech, many of the nerve elements are 
paired: a large element in the right half of the ganglion has 
its fellow, similar in size, form and extent, symmetrically placed 
on the other side. More interesting still is the fact that doth of 
these paired elements usually take the stain together, indicating the 
existence of connection between them. In many of my 
preparations of the second abdominal ganglion a pair of large 
motor elements was often demonstrated. One of these from 
the left side of the ganglion is shown in Figure 10 (Plate VI) ; 
its fellow of the right side was its mirrored image even to the 
number, position and extent of the collaterals. The collaterals 
always branch to the same points in the neuropil, a fact directly 
against the assumption that a diffuse fibrillar network exists. 
For if this were the case, why should the processes of different 
neurones always pass to the same spot in the neuropil, and 
why should the nerve elements be arranged in bilateral sym- 
metry? Such an arrangement would be useless if there were a 
diffuse network of fibrillae in the neuropil, to put the dendrites 
of all the nerve elements into general communication with one 
another. 

Figure 10 also illustrates a second point, which, as we shall 
see, bears upon the fibrillar structures in the nerve elements. 
The nerve fiber reaches its greatest size in the region desig- 
nated by a, and this ts the point at which most of the collaterals 


PRENTISS, Weurofibrillar Structures. 167 


enter or leave the neurone; both the cell process (b) and the 
peripheral fiber (c) ave considerably smaller than the region inter- 
vening between them. ‘The greater size of this particular por- 
tion of the elements is explained when the fibrillar structures are 
studied under a higher magnification (Figure 11, Plate VI). It 
may then be observed that the fibrillae are more numerous at this 
point than in either the cell process proper, or in the peripheral 
fiber. Many fibrillae of the large collaterals (d’) pass directly to 
the periphery through the fiber c’; others, like fibril e, evidently 
enter the element through one collateral and pass out through 
another. The same condition was observed in the nerve 
elements of the leech (Figure 2, Plate V). The figure is from 
a preparation fixed in ether and stained with toluidin blue. 
The plane of section was very favorable, showing the whole of 
the cell process and short portions of the peripheral fiber and of 
two collaterals in connection with the cell. Here again an 
enlargement is found at the point where the large collaterals 
branch off. As only fibrillar structures were stained, each 
fibril could be traced with perfect distinctness; a large one (a’) 
passes directly from a collateral at @ into the peripheral fiber 6 
and is, therefore, independent of the ganglion cell. 

In the large nerve elements of the leech a fibrillar net- 
work is often found in that portion of the cell process from 
which the collaterals branch. Two such cases have been 
figured (Figures 5 and 7, Plate V). Figure 5 isa type of the 
more simple connections which exist between the fibrillae in the 
nerve elements. Three large parallel fibrillae (a, a’, a’) unite 
at the point 0; atc, a single mesh is formed, from which are 
given off several smaller fibrils that continue their parallel 
courses in the process. In this, as in all well differentiated 
molybdate preparations, the perifibrillar substance is not 
stained, and the boundaries of the nerve elements are indicated 
by the course of the fibrillae only. 

The processes of the giant ganglion cells of the leech 
often exhibit extremely complicated fibrillar networks soon 
after they enter the neuropil (Figure 7, Plate V). In Figure 7 
three large fibrillae from the longitudinal commissure unite 


168 JournaL oF CompaRaTiveE NEUROLOGY. 


with the process;.one of these fibrillae (d) joms a small fine- 
meshed network at ¢; the fibrillae coming from the cell at 2 are 
all put into communication with one another soon after they 
enter the neuropil by the network at « Such connecting net- 
works form direct paths for nervous impulses passing from one 
merve element to another, and im entire independence of the 
ganglion cells. These networks have been observed only m 
the processes of the larger cells. Similar fibrillar structures 
have been seen by Berse, whose observations have not as yet 
been published The large single fibrillae which unite with the 
fibrils of the process at right angles often divide forming 
T-shaped branches. Im one case such a fibril was traced into 
direct connection with the process of another nerve element 
(Prentiss, : 03, Figure I9)- 

Figure 1 (Plate V) shows an imteresting case of fibrillar 
continuity within the process of a giant ganglion cell. The 
process is so sectioned as to show only a portion of its cell, but 
may be traced into the neuropil together with the three large 
neurofibrillae which it contains. Immediately within the neu- 
ropil, it is jommed- by two longitudinal fibrillae (2 and 4), which 
branch and unite with two of the fibrillae from the cell; a cross 
fibril, c, puts into direct connection with each other a and &, 
which are fibrillae from the longitudinal commissures and are 


evidently in connection with other ganglia, lying anterior and- 


posterior. A nervous impulse, therefore, which was transmitted 
from the ganglion next anterior to this one, or even from the 
brain, would be conducted through fibril a, stimulate the nerve 
element, and pass at once through cand 4 to the nerve elements 
in the next posterior ganglion. I have never observed such 
commissural fibers branching dichotomously like the dendrites 
of motor elements, but they may constantly be seen in connec- 
tion with cell processes, as in this case. The fibrillar structures 
shown in Figure 1 were stained as opaquely as the fibers in 
Gols! preparations, and since they were situated in a perfectly 
clear space, the connections between the various fibrils were, 
under a magnification of 2000 diameters, as distinct as a dia- 
gram. The further continuations of the fibrillae in the cell 


ee ee ee 


———————E 


Prentiss, Weurofiirillar Structures. 169 


process were obscured by longitudinal fibrillae; they cross irreg- 
ularly, but whetker a network was formed could not be deter- 
mined with certainty. 

In the crayfish was found one case of apparent connection 
between the dendrite of a large motor neurone and a longitud- 
inal connective fiber (Figure 12, Plate VI). To economize 
space only a portion of the collateral 2 is shown im the figure, 
but in the preparation this collateral was traced into connection 
with a motor ganglion cell 46, one of the end branches of the 
collateral 2, may be observed uniting with the small longitud- 
inal fiber dat the pointc. This fiber contains several neuro 
fibrillae, while 4, if not a single fibril, is composed of not more 
than two. 4 does not branch on uniting with a, but can be 
traced a short distance toward ¢, the anterior end of the longi- 
tudinal connective. In the preparation, d was seen to extend 
in both directions beyond the limits of the ganglion; naturally, 
its cell was not demonstrated, but the fact that it is much larger 
than 6 is sufficient ground for believing that all of the neuro 
fibrillae which it contains do not come from the nerve element 
a. The connection between 4 and dat the pomt ¢ seems as 
Certain as if they were the branches of a single neurone. 

Fibrillar Netwerks mm the Neuropil. 

A considerable number of cases of fibrillar networks occur- 
ring in the neuropil of Hzude have been described in another 
paper (PRENTisS, : 03); two additional examples of the types 
which occur most frequently in Hirvude, are figured here. Often 
only two or three meshes are formed with which two or three 
fibrillae are connected (Figure 6, Plate V). In other cases the 
networks are much more complex, as seen in Figure 8. These 
examples were found near the center of the neuropil, and there 
is little likelihood of their being mistaken for networks m the 
cell processes. In every instance they were of lmited extent, 
and comparatively few fibrillae were connected with them. 

In Astacus more extensive networks may be demonstrated 
in the neuropil by means of methylen blue. Figures 9 and 13 
are good examples of these fibrillar structures. The meshes 


170 JOURNAL OF COMPARATIVE NEUROLOGY. 


are larger than in the leech, and beads of perifibrillar substance 
are scattered along the course of the fibrillae, often at their 
points of union. These networks correspond more nearly to 
those which ApATHy demonstrated in the leech with methylen 
blue, but they are not so extensive. 

My observations thus confirm the statements of APATHY 
that fibrillar networks occur in both ganglion cells and neuropil. 
The evidence, however, does not support his view that the 
networks within the cells are formed by neurofibrillae which 
differ from each other in both structure and function. This 
conclusion is, moreover, in entire agreement with BETHE’s ex- 
periment which proved that the cells are not the centers of ner- 
vous activity as APATHY supposed. As to APATHY’S assump- 
tion of a diffuse fibrillar network in the neuropil, there is no 
evidence to show that such a condition exists. There are rather 
numerous small networks, each limited to a definite region in 
the neuropil, and putting comparatively few fibrillae into com- 
munication with one another. 

Fibrillar structures in my preparations of both H/zsudo and 
Astacus, confirm BETHE’s statement that many neurofibrillae are 
found in the nerve elements which are entirely independent of 
the ganglion cells. In A/zvudo all the neurofibrillae of a nerve 
fiber may be put into communication with one another by net- 
works before they enter the cell. 

These facts all point to the existence of fibrillar continuity 
between the nerve elements; in both H/zvudo and Astacus appar- 
ent cases of continuity between two neurones have been ob- 
served. Great weight cannot be laid on only two observations 
of such connection, for there is always the possible danger of 
optical error in tracing such exceedingly small structures. But 
the very existence of independent neurofibrillae in the nerve 
elements and the presence of fibrillar networks in the neuropil 
are incompatible with the idea that the nervous system is com- 
posed of anatomically independent, cellular units. On the con- 
trary such conditions can be explained only by assuming that 
between the nerve elements fibrillar continuity exists. 

The proof of such continuity does not, however, annihi- 


PrENTIss, Neurofibrillar Structures. 17E 


late the neurone theory ; it simply modifies it. There is as yet 
_no direct evidence to prove that these fibrillar networks in the 
neuropil may not be formed by the union of fibrils, each of 
which was developed in a distinct cell. At present we know 
nothing of the origin of the neurofibrillae in the central ner- 
vous system, and, therefore, the presence of independent fibril- 
lae in the nerve elements does not necessitate the abandonment 
of the neurone idea; for such fibrillae, which do not enter the 
cell, may be differentiated in the plasma of the cell processes, 
and be as much a part of the cell as the processes themselves. 
The fundamental experiment of Berne (’98) while of great 
value to neurology, does not in itself prove the neurone theory 
false ; it merely shows that the nerve elements may function for 
some time without their cells; that the cells are not the batteries. 
which generate the nervous current, as was formerly supposed. 

An unprejudiced thinker can but agree with VERWORN 
(: 00) when he says that: ‘‘der Begriff des Neurons und damit 
auch die Neuronenlehre erst dann und nur dann erschittert 
ware, wenn es gelungen, zu zeigen, dass das, was wir als cellu- 
lare Einheit betrachten, in Wirklichkeit aus mehreren Zellen 
besteht.’”’ The only important evidence which at present goes 
to show that the ‘‘cellulare Einheit’’ of the nervous system 
‘‘aus mehreren Zellen besteht,’’ consists of BETHE’s preliminary 
statements of his researches in the histogenesis and pathology 
of the nerve elements. Before passing judgment on BETHE’s 
work we must wait until a full account of it appears. If, as is 
probable, the results of his work prove beyond a doubt that the 
supposed cell individual is the product of many cells, the neu- 
rone theory will be untenable. If not, the neurone theory will 
still remain—a ¢heory; for our present methods have thus far 
failed to prove ‘‘beyond a doubt,” that the nervous system is 
made up of cell units, and nothing but cell units. 


172 JOURNAL OF COMPARATIVE NEUROLOGY. 


LITERATURE. 


APATHY, S. 
’97. Das leitende Element des Nervensystems und seine topographis- 
chen Beziehungen zu den Zellen. Mitth. d. zooloy. Station z. Neapel. 
Bd. 12, p. 495-748, Taf. 23-32. 
BALFour, F. M. : 
*76. On the Development of the Spinal Nerves in Elasmobranch Fishes. 
Phil. Trans. Roy. Soc. London, Vol. 166, p. 175-199, pls. 16-18. 
BETHE, A. 
’98. Das Nervensystem von Carcinus maenas. Arch. f. mikr. Anat. 
Bd. 51, p. 382-452, Taf. 16-17. 
BETHE, A. 
98a. Ueber die Primitivfibrillen in den Ganglienzellen von Menschen und 
andern Wirbelthieren. Morph. Arbetten, Bd. 8, p. 95-116, Taf. 9-10. 
BETHE, A. 
:00. Ueber die Neurofibrillen in den ganglienzellen von Wirbelthieren 
und ihre Beziehungen zu den Golginetzen. Avch. f. mtkr. Anat. 
Bd. 55, p- 513-558, Taf. 29-31. 
BETHE, A. 
:00a. Das Molybdanverfahren zur Darstellung der Neurofibrillen 
und Golginetze im Centralnervensystem. Zettschr. f. wiss. Mikrosk. 
deity pape 13-35). 
BETHE, A. 
:02. Ueber die Regeneration peripherischer Nerven. Arch. f. Psychiatr. 
Bd. 34, p. 1066-1073. 
DoHRN, A. 
’o1. Nervenfaser und Ganglienzelle. Histogenetische Untersuchungen. 
Mitth. d. zoolog. Station z. Neapel. Bd. 10, p. 225-341, Taf. 16-22. 
FoREL, A. ; 
86. Einige himnanatomische Betrachtungen und Ergebnisse. Arch. /. 
Psychiatr. Bd. 18, 37 pp. 
GEHUCHTEN, A. VAN 
:00. Anatomie du systéme nerveux de l’homme. Legons professées a 
Université de Louvain, 3me édition, Louvain, 2 vols., 1106 pp., 
702 figs. 
GUDDEN, B. VON 
’*89. Gesammelte und hinterlassene Abhandlungen. Herausgegeben von 
H. Grashey. Wiesbaden, 6 + 221 pp., 41 Taf. 
HARRISON, R. G. 
:o1. Ueber die Histogenese des peripheren Nervensystems bei Salmo 
salar. Arch. f. mtkr. Anat. Bd. 57, p. 353-444, 7 Fig., Tafel 18-20. 
His, W. 
’89. Die Neuroblasten und deren Entstehung im embryonalen Mark. 
Arch. f. Anat. u. Physiol. anat. Abtheil. p. 249-300, Taf. 16-19. 
His, W. 
’90. Histogenese und Zusammenlaug der Nervenelemente. Arch. f- 
Anat. u. Physiol. anat. Abthetl., Suppl. Bd., p. 97-117, 30 Fig. 


Prentiss, Meurofibrillar Structures. 173 


HocuHeE, A. 

’99. Der gegenwiartige Stand der Neuronenlehre. eurol. Centraldl. 

18 Jahrg., 37 pp- 
LENHOSSEK, M. VON.: 

99. Kritisches Referat iiber die Arbeit A. BETHE’s: Die anatomischen 
Elemente des Nervensystems und ihre physiologische Bedeutung. 
Neurol. Centraibl. 18 Jahrg., p. 242-246, 301-308. 

MEYER, S. 

99. Ueber centrale Neuritenendigungen. Arch. f. mikr. Anat. Bd. 54, 

p. 296-311, Taf. 17. 
MONCKEBERG, G., und BETHE, A. 

?99. Die Degeneration der markhaltigen Nervenfasern der Wirbelthiere 
unter hauptsichlicher Berticksichtigung des Verhaltnisses der Primi- 
tivfibrillen. Arch. f. mikr. Anat. Bd. 54, p. 135-189, Taf. 8-9. 

NIss.L, F. 

:03. Die Neuronenlehre und ihre Anhaénger—ein Beitrag zur Lésung 
des Problems der Beziehungen zwischen Nervenzelle, Faser und 
Grau. Jena. vi- 478 pp., 2 Taf. 

OWSIANNIKOw, P. 

99. Ueber die Nervenelemente und das Nervensystem des Flusskrebses, 
Astacus fluviatilis. Mém. Acad, St. Petersburg. (8) Tome to, 
30 pp., I Tafel. 

PRENTIss, C. W. 

:03. Ueber die Fibrillengitter in dem Neuropil des Hirudo und Astacus 
und ihre Beziehung zu den sogenannten Neuronen. Archiv f. mtkr. 
Anat. Bd. 62, pp. 592-606, Taf. 26. 

VERWORN, M. ‘ 
:00. Das Neuronin Anatomie und Physiologie. /ena. 54 pp., 22 Fig. 
WALDEYER, W. 

’91. Ueber einige neue Forschungen im Gebiet der Anatomie des Ner- 

vensystems. Deutsche med. Wochenschrift. No. 44-50. 


EXPLANATION OF PLATES. 


The fibrillar structures which are described in the text are repro - 
duced by dark lines, as they appeared in the preparations; the out- 
lines of the less important fibrils were drawn much lighter, for the 
sake of clearness. The Figures were projected and outlined with a 
Leitz camera lucida (ABBE model). With the exception of Figure ro 
the drawings were made with the use of a LeIrz 1-16 oil immersion 
objective, and a No. 4 ocular. With a tube length of 160 mm. and a 
projection distance of 320 mm. this system gives a magnification of 
approximately 2000 diameters. Figure 10 is magnified 160 diameters 
by means of a LeITz objective No. 3, and ocular No. 4. 


174 JOURNAL OF COMPARATIVE NEUROLOGY. 


PLATE V. 


All the figures are from toluidin blue preparations of Azrudo. Figures 2 
and 3 are from material fixed in ether. 

Fig. r. An oblique section through a large ganglion cell and a portion of 
the neuropil, showing the connection between the fibrillae in the cell process. 
The greater part of the cell is cut away. a, 4,two fibrillae from the longi- 
tudinal commissures; ¢c,a short fibril connecting @ and 4; d, the bounding 
membrane between neuropil and ganglion cells. 

Fig 2. A-small ganglion cell with its process, portions of the peripheral 
fiber (4), and of two collaterals. Part of the peripheral fibrillar network is seen 
in the cell; in that region of the element, from which the collaterals branch 
off, a large fibril (a) may be traced from one of the collaterals into 4, the 
peripheral fiber, where it is designated by a’. 

Fig. 3. A ganglion cell of medium size from the same section as Figure 
2. Only the neurofibrillae and the nucleus were visible in the preparation. 4, 
the single large fibril in the cell process ; a, one of its branches which unites 
with the network about the nucleus. 

Fig. g. A section through the peripheral portion of a giant ganglion cell, 
showing the close-meshed fibrillar network characteristic of the larger cells of 
Hirudo. 

Fig. 5. Ashort portion of a cell process showing the connection between 
its neurofibrillae ; a, a’, a’’, three fibrillae which in passing from the cell into 
the neuropil, unite at 6; atc asingle mesh is formed, from which numerous 
small fibrillae continue their course in the process. The perifibrillar substance 
is not stained. 

Fig. 6. One of the simpler cases of fibrillar networks found in the 
neuropil of Azrudo. Only two meshes are formed, which are in connection 
with three fibrillae. 

Fig. 7. An example of the fibrillar networks in the processes of large 
nerve elements. a, proximal end of the cell process; 4, bounding membrane 
of the neuropil; ¢, the network by which all of the fibrillae are connected 
together ; d, a large neurofibril which unites at e with a smaller network. 

Fig. 8. A fibrillar network of four meshes found near the center of the 
neuropil. a-/, fibrillae uniting with the network. 


PLATE VI. 


Figure 14 is a toluidin blue preparation from Airudo. All the other fig- 
ures are methylen blue preparations from Ast¢acus material. 

Fig. 9. Anetwork of neurofibrillae from the third abdominal ganglion 
of Astacus. At the point a the network is apparently connected with the longi- 
tudinal nerve fiber 4; c, beads of perifibrillar substance, characteristic of 
methylen blue preparations. 

Fig. ro. <A large motor nerve element from the second abdominal 
ganglion of Astacus. a, the enlarged portion of the fiber from which the 
collaterals branch off; 4, the much smaller cell process; ¢, the peripheral 
nerve fiber, traced, in the preparation, to one of the lateral nerves; d, two 
collaterals. XX 160. 


PLATE V, 
JOURNAL OF COMPARATIVE NEUROLOGY. VoL. XIII. 


— 


——— 


Cc. W. P. DEL. 


PLATE VI. 


JOURNAL OF COMPARATIVE NEUROLOGY. VOL. XIil. 


F 


] 


SES LS A I NOTE sa TT 


+, Boston, 


Heliotype Co 


ieee: Sse eae ; Low, - 
‘ 2 ea eee hie Se ee : ‘ 
“ é bi wi) 
iy san 
- | oA 4 
dt r n 
yi - 7 ay 
Ay +5 
APY 
fee 
eo ; 
* f : 
4 : 
, ¥ Pha ~s , 
i ee 
ve hase 
. 


PRENTISS, (Veurofibrillar Structures. 175 


Fig. rr. An enlarged drawing. of that portion of Figure 10 from which 
the collaterals are given off. 0’, c’, a’, correspond respectively to 4, c, din 
Figure 10. Many of the fibrillae do not enter the cell process; of these, ¢ 
passes directly from one collateral to another. 

Fig. 12. The collateral of a large nerve element which is apparently in 
direct connection with a longitudinal fiber. a, the collateral ; 4, the connect- 
ing fibril; c, its point of union with the fiber d; ¢, the anterior end of the 
longitudinal fiber d. 

Fig. 73. A fibrillar network from the neuropil of the third abdominal 
ganglion; a-f, neurofibrillae which are connected with the network; g, beads 
of perifibrillar substance. 

Fig. rg. One of the smaller ganglion cells of Airvudo (motor type of 
APATHY), showing the inner and outer fibrillar networks, which are put into 
connection by radial fibrillae. 

Fig. 15. A ganglion cell from the last abdominal ganglion of Astacus. 
Only the fibrillae near the periphery of the upper half of the cell are shown. 


ON THE INCREASE IN THE NUMBER OF MEDUL- 
PED NERVE oPIBERS IN THE “VENTRAL 
ROOTS OF THE SPINAL NERVES OF THE GROW- 
ING WHITE RAT. 


By SHINKISHI Hartat. 


(from the Neurological Laboratory of the University of Chicago) 


This investigation is a continuation of previous work (’02) 
on the medullated fibers in the dorsal roots of the spinal nerves. 
The following results were obtained from the earlier investi- 
gation: 

The spinal ganglia of the white rat were found to contain 
two kinds of cells which are differently characterized. One 
kind is of large size, while the other is much smaller. From 
the histological evidence, the larger cells were regarded as the 
adult functional form, while the smaller were considered to be 
the cells in a developing stage (’o1). 

If this view is correct and if the total number of cells in 
the ganglion remains the same throughout life, then we should 
expect more of the large cells in the ganglion of the adult 
than in that of the young animal. The converse should also 
be true; the younger animal should have more of the smaller 
cells in the ganglia. By counting the ganglion cells in animals 
of different ages, this hypothesis can be tested. According to 
BUHLER (’98), the large functional cells are constantly degen- 
erating, and the smaller cells which are developing take their 
place. If his view were true, the spinal ganglion cells should 
decrease in number as the animal gets older, since it is believed 
that in the rat the ganglion cells never divide after the first 
days of extra-uterine life. BiHLEer’s hypothesis, therefore, 
can be tested by an enumeration of the total number of the 
spinal ganglion cells at different ages. For these two reasons, 
I undertook this investigation. 


178 JOURNAL OF CoMPARATIVE NEUROLOGY. 


I enumerated the total number of the spinal ganglion 
cells and dorsal root fibers at different ages in several spinal 
nerves of the rat, and arrived at the following conclusions( ’02): 

1. The total number of the spinal ganglion cells remains 
approximately constant between 10.3 grams (age 10 days) and 
167 grams of body-weight (age 100 to 200 days). 

2. All of the small cells contained in the spinal ganglia 
are constantly growing, and some of them become large cells. 
So that the number of large cells increases with age, while 
there is a corresponding decrease in the number of small cells. 

3. The number of the cells in the spinal ganglia is 
always more than twice the number of medullated fibers in the 
corresponding dorsal nerve roots. 

From the summary it is evident that the small cells are in 
a developing state and are becoming larger, while the fact that 
there is no numerical decrease in the total number of the cells 
favors my view, and is against that of BUHLER. 

In the present investigation the observation has been 
extended to the ventral root fibers. The method of prepara- 
tion and the technique of counting were the same as used in 
the earlier investigation (’02). The following are the main 
results from this latter study: 


TABLE I. 


The Total Number of the Medullated Ventral Root Fibers in the 
Spinal Nerves Here Named. 


Weight of Rat 10.3 grms. 25.4 68.0 164.9 264.3 
Z2e2 
Gy OME Gers. 558 1007 1302 1474 1522 | 225 
Sele bor. 286 434 561 613 772 \) Cneics 
2 1 II Lumb. 333 698 704 1028 965 De 
| Bee 
SEEN IES 


As we see from Table I, the number of ventral root fibers 
steadily increases from 10.3 to 264.3 grams. The relative 
increase is nearly the same for the different nerves examined. 
Generally speaking, the total number of the medullated fibers 
in the adult animal is approximately 2.7 times that in the 10.3 
gram rat, and the fibers increase most rapidly between the 


Hartal, Spenal Nerves of the Rat. 179 


weight of 10.3 and 25.4 grams, after which the increase is 
slower. In this respect the changes in the ventral roots are 
similar to those found in the dorsal roots. 

Comparison of the number of the dorsal root fibers at 
10.3 and 264.3 grams’ indicates nearly the same rate of 
increase as is found for the ventral roots. The following table 
shows this : 


TRABEE lle 
Relative Increase of the Medullated Dorsal and Ventral Root Fibers. 


Weight of VI Cervical IV Thoracic II Lumbar 
Rat Dorsal Ventral Dorsal Ventral Dorsal Ventral 
10.3 I I I I I I 

264.3 2.0 a7 2.9 Pela] 2.9 2.8 


As the table indicates, the total number of the dorsal root 
fibers of the 264.3 gram rat is about twice that of the I0.3 
gram rat in the cervical region, while in the thoracic and 
lumbar regions the fibers are nearly three times as numerous. 
The different ratio shown between the three regions is probably 
due to the fact that the number of dorsal root fibers in the 
cervical region increases more rapidly between birth and 10.3 
grams, than in the case of the two other regions. This special 
relation, however, should be further investigated. 

A comparison between the number of the ventral and 
dorsal root fibers presents an interesting relation. 


TABLE III. 


Comparison Between the Number of the Medullated Ventral and 
Dorsal Root Fibers at Different Ages. 


Weight of Rat. 10.3 25-4 68.0 164.9 264.3 
Root Wentia Dorssam Vin Dies SVin el): Vi» Vie Ds 
VI Cerv. 558 Ig98 1007 2569 1302 3683 1474 4227 1522 4028 
IV Thor. 286 607 434 863 561 1420 613 1522 772 1650 


II Lumb. 333 723 698 9Q11 704 1317 1028 1644 965 2102 
Total Number ~=1177 3328 2139 4343 2567 6420 3115 7393 3259 7780 


Ratios I PCO Ab 2 I Pats i at) | PERU Vee Mpa. 


1The observation on this animal has been made since the previous work 
('02) was published. 


180 JOURNAL OF COMPARATIVE NEUROLOGY. 


For clearness, only the totals obtained by adding the 
fibers in the roots as shown in Table III, will be discussed. 

As we see from the above table, the ratios between the 
two roots in 10.3 grams is 1:2.9, while in the mature animal it 
is 1:2.3. This means that the number of the ventral root fibers 
increases more rapidly than that of the dorsal root fibers, con- 
trary to what appears in the frog (BirGE, 1882). The addition 
of the new fibers between 165 and 264 grams is very small, 
and therefore the animal having a body weight of 165 grams 
may be regarded for this purpose as adult. These results 
show that the number of medullated fibers in the ventral roots 
of the spinal nerves of the rat as well asin the dorsal roots 
increase more than 100% with age—the age being indicated 
here by the body-weight. On the other hand, ScHILLeEr (’89), 
who counted the number of the medullated fibers in the oculo- 
motor nerve of the cat at different ages, showed that the nerve 
of the new-born kitten contained 2942 fibers, while that of the 
eighteen months old cat was found to have 3035 fibers. 

Thus, according to him, the number of the fibers added 
between birth and maturity was only 83; or hardly 3% more 
than the number found at birth. If this observation is correct, 
it indicates that the development of these efferent neurones in 
the cat is completed very early. The lack of accordance in the 
observations on the oculomotor nerve of the cat with those on 
the efferent spinal nerves of the rat, demands further study. 

Several years ago, Harpesty (’99), made the important 
observation on the spinal nerves of the frog that the number of 
medullated dorsal root fibers is more numerous near the 
ganglion than at the entrance of the dorsal root into the cord, 
while in the case of the ventral root, the number of medullated 
fibers diminishes from the cord towards the ganglion. This he 
interpreted to mean that the dorsal as well as ventral root fibers 
are growing (medullating) even in an older animal, and there- 
fore as we pass away from the cells of origin, the number of 
mature fibers diminishes. DaLe (’00), who counted the num- 
ber of medullated fibers in the dorsal roots of several spinal 


Hatai, Spenal Nerves of the Rat. 181 


nerves in the adult cat, was not able to observe in that animal 
the numerical differences recorded by Harpesty for the frog. 

I have pointed out already, that the total number of the 
medullated fibers in the two roots increases from 10.3 to 264.3 
grams, although the increase is not so marked between 65 and 
264.3 grams as at the younger ages. From these facts, one 
might expect at the earlier ages an excess of fibers near the 
ganglia in the case of the dorsal roots, and near the cord in the 
case of the ventral roots, since the fibers are outgrowths of the 
cell-body located in spinal ganglion and ventral horn, respec- 
tively. In order to test this point, the fibers were enumerated 
at two levels in the ventral roots, one near the cord and the 
other farther from the cord and at the point where the ventral 
root approaches close to the ganglion, as had been done by 
Haropesty in the frog, and the following results were obtained : 


TABLE IV. 


Numerical Difference of the Medullated Ventral Root Fibers of all 
three nerves, VI Cerv., IV Thor., and IJ Lumb., combined at 
the Two Levels, Proximal (near cord) and Distal (near ganglion). 


Weight of 

Rat 10.3 ZEA: 68.0 164.9 264.3 
Proximal 1299 2244 2696 3169 3279 ) Number of the 
Distal 7 2139 2567 3095 3259 | medallted 
Ditference 122 105 129 74 20 

or 9-4% 4.6% 47% 2.3% 0.6% 


As we see from the table, the ventral root in the proximal 
section, i. e., near the cord, contains more medullated fibers 
than appear in amore distal section. This difference at the 
two levels is greatest in the youngest animal (122 fibers or 
9-4% in excess at 10.3 grams) and diminishes to 0.6% in the 
oldest rat examined. Since the difference decreases as the 
animal becomes older, we most readily explain the decrease of 
the difference by the growth of the fibers which at an earlier 
age are immature, and therefore stop short of the more distal 
section. We infer that even when the growth process stops, 
there are some fibers which are still incomplete and unfinished ; 
this explains the fact that in the oldest rats differences are found 


182 JOURNAL OF COMPARATIVE NEUROLOGY. 


between the proximal and distal sections, although at this phase 
of the life cycle, the difference does not necessarily mean that 
crowth in these rats is still taking place. Thus Harbesty’s 
observations on the frog are confirmed in the growing white rat. 


Summary. 


1. The total number of the medullated fibers in the 
ventral roots of the spinal nerves increases as the animal be- 
comes older (Table I). The rate of increase of these fibers is 
not the same for the different ages. It is most rapid between 
the weights of 10.3 and 25.4 grams (10-30 days), after which 
it becomes slower. The number at maturity is about 2.7 
times that found in the 10.3 gram rat. 

2. The total number of the medullated fibers in the 
ventral roots in the 10.3 gram rat is approximately one-third of 
that in the dorsal root fibers (1:2.9), while in the adult the 
ratio is 1:2.3 (Tables IJ and III). Thus the increase of 
medullated fibers in the ventral roots is more rapid than in the 
dorsal roots. 

3. At all ages, the ventral root near the cord contains 
more medullated fibers than appear more distally, the second 
section being taken near the ganglion of the corresponding 
dorsal root; in other words, the total number of the medul- 
lated fibers in the ventral root decreases from the spinal cord 
towards the periphery (Table IV). 


BIBLIOGRAPHY. 
BIRGE, E. A. 
82. Die Zahl der Nervenfasern und der motorischen Ganglienzellen im 


Riickenmark des Frosches. Arch. f. Anat. u. Phystol., Physiol. 
Abtheil. Hefte 5 u. 6. 


BUHLER, A. 
’98. Untersuchungen iiber den Bau der Nervenzellen. Verhandlungen 
der Physik. med. Gesellschaft zu Wiirzburg. N.F., Bd. 31. 
DALE, H. H. 
oo. On some Numerical Comparisons of the Centripetal and Centri- 


fugal Medullated Nerve Fibers Arising in the Spinal Ganglia of the 
Mammal. /ourn. of Physiol. (Foster), Vol. XXV, No. 3. 


Se ee 


Hatal, Spenal Nerves of the Rat. 183 


HARDESTY, I. 
’99. The Number and Arrangement of the Fibers Forming the Spinal 
Nerves of the Frog (Rana virescens). /ourn. Comp. Neurol. 
Vol. IX, No. 3. 


HARDEsTY, I. 
’oo. Further Observations on the Conditions Determining the Number 
and Arrangement of the Fibers Forming the Spinal Nerves of the 
Frog (Rana virescens). Journ. Comp. Neurol. Vol X, No. 3. 


HATAI, S. 
?o1. The Finer Structure of the Spinal Ganglion Cells in the White 
Rat. Journ. Comp. Neurol. Vol. XI, No. t. 


EVA TAT, Os 
’02. Number and Size of the Spinal Ganglion Cells and Dorsal Root 
Fibers in the White Rat at Different Ages. Journ. Comp. Neurol. 
Vole SUH No: 2. ; 


SCHILLER, 
’89. Sur le nombre et le calibre des fibres nerveuses du nerf oculomo- 
teur commun, chez le chat nouveau-ne et chez le chat adulte. 
Compt. Rend. Acad. d. Sct., Parts. 


ON THE MEDULLATED NERVE FIBERS CROSSING 
tobe star Or, LESIONS, IN, THE) BRAIN) OF 
THe WHITE RAT. 


By S. WALTER Ranson. 
Instructor in Anatomy, Marion—Sims— Beaumont Medical School St. Louis University 


(From the Neurological Laboratory of the University of Chicago.) 


With Plate VII. 


I. Htstorical Review. 


The results of research on the regeneration of the neu- 
rones in the central nervous system of vertebrates have varied 
greatly according to the age and zoological position of the 
animals used for experiment. For this reason the observations 
found in the literature have been grouped for review according 
as they were made on amphibians, reptiles, birds, or mammals. 
It should be born in mind in considering this question, that 
regeneration in the central nervous system may involve either a 
proliferation of neurones as a whole, 1. e., the growth of new 
elements resulting from cell division, ora restitution of parts of 
mature neurones which have been mutilated. 


Spinal Cord of Amphibia. 


In the spinal cord of amphibia the process has been de- 
scribed quite fully by BarrurtH (2) (1891), who amputated 
the tails of frog larvae and was able to show that the cells 
lining the central canal became amoeboid and arranged them- 
selves so as to close the canal. These cells then proliferated 
by indirect division, forming a tube which extended backward 
into the new tail. From this tube the new spinal cord with its 
ganglia and nerves developed. FRralissE (12) (1885) found a 
normally functioning cord in the regenerated tail of an amphib- 
ian larva. Harrison (16) (1898) secured results agreeing with 


186 JOURNAL OF COMPARATIVE NEUROLOGY. 


those of previous investigators, except that the spinal ganglia 
regenerated only in the most proximal part of the new-formed 
cord. Scosso (29) (1891) obtained positive results on frog 
larvae and tritons although the spinal cords of other vertebrates 
gave no evidence of regeneration. 

The cases of the adult. salamander and triton are very 
similar. Caporaso (6) (1887), working with the latter animal, 
found that after cutting off part of the tail the spinal cord 
degenerated cephalad for a short distance, the cells lining the 
central canal remaining intact. These ependyma cells then 
proliferated, restoring first the degenerated area cephalad to the 
injury, and then extending backward to restore the part lost by 
the amputation of the tail. Within the new formed mass 
ganglion cells were differentiated. Scosso (29) and BARFURTH 
(2) have also reported positive results on the triton. MULLER 
(26) (1864) observed in the regenerating tail of the salamander 
a new cord, which resembled in appearance the lost part, but 
which did not establish functional connection with the new tail. 

_ The results of work on the frog are much less conclusive. 
Mastus and VANLAIR (24) (1869), after removing 2 mm. of the 
spinal cord of a frog, found that the cavity was subsequently 
filled with tissue containing what they took to be nerve cells and 
nerve fibers. Later Masius (23) (1880) reasserted his former 
conclusions, butas neither Scopso (29) (1891) nor MaRINESCO 
(22) (1894) have been able to confirm his observations, it seems 
doubtful whether the spinal cord of the adult frog possesses 
even this degree of regenerative capacity. 

In brief, all amphibian larvae and the adult triton can 
perfectly restore a lost portion of the spinal cord; the sala- 
mander accomplishes this only incompletely; and it is an open 
question whether the frog can do so at all. 


Spinal Cord of Reptiles. 

Lizards have been studied by MU ier (26) (1864) and 
FRalIssE (12) (1885), each of whom found the lost portion of 
the spinal cord partially reformed, but not so completely as to 
be of functional value. On the other hand, Scosso (29) (1891) 


Ranson, F2bers 12 Lestons of the Brain. 187 


was unable to find any evidence of a regeneration of the spinal 
cord in these animals. 


Spinal Cord of Birds. 


Upon birds we have the observations of BROwWN-SEQUARD 
(4) (1851), who severed the spinal cord in pigeons and subse- 
quently found nerve cells and nerve fibers in the tissue joining 
the cut ends. Ina paper published in 1892 he reaffirms his 
conclusions without further experimentation (5). Scosso (29) 
(1891) also experimented with birds, but failed to secure any 
evidence of regeneration. The observations on reptiles and 
birds are so few in number and so contradictory in character 
that no definite statement can be made. A further study of 
these forms would be desirable. 


Brain of Frogs and Birds. 


Much less work has been done on the brains of amphibia, 
reptiles, and birds. Two researches have been recorded in this. 
field, each giving positive results, though neither has been con- 
firmed. DaNIELEWsky (9) (1891) completely removed the 
cerebral hemispheres of a frog; and after nine months found a 
new cerebral mass, a microscopic study of which revealed cells 
which were believed to be nerve cells. Von Voir (35) (1868) 
claims to have secured even more striking results on the 
pigeon. Five months after complete removal of the cerebral 
hemispheres in a young pigeon he found two new hemispherical 
masses, with a cavity in each. These masses, which were 
composed of nerve fibers and nerve cells, passed over into the 
crura cerebri. There has been no serious attempt to repeat 
these observations and they have neither been confirmed nor 
disproven. It would be profitable to apply to the study of the 
brains of the lower vertebrates the same methods that have 
been employed in the study of wounds in the brains of mam- 
mals, since there is nothing in this part of the literature to 
compare with the careful studies of CoEN, SANARELLI, TEDESCHI 
and TscHISTOWITSCH upon the mammalian brain. 


188 JOURNAL OF COMPARATIVE NEUROLOGY. 


Spinal Cord of Mammals. 


It will perhaps be noticed that in this review the question 
of the restoration of function has been carefully avoided. This 
is because we are interested here only in the presence or ab- 
sence of new-formed nerve elements, and this is a matter which 
can be studied quite apart from the physiological value of these 
elements. This restriction will make possible a clearer view of 
the literature on the regeneration of the spinal cord of mam- 
mals, which has been much confused by the failure to appre- 
ciate the difference between reflex and voluntary movements. 

Eicuorst and Naunyn (10) (1874) crushed the spinal cord 
of dogs. In the connective tissue which developed in this 
region many nerve fibers were found. These resembled the 
fibers of peripheral nerves, from which Ercuorst believes they 
are ingrowths. In one instance he was able to follow one of 
these fibers to a spinal ganglion. Since it was impossible to 
prove that all the fibers had been destroyed by crushing, these 
results were not favorably received. To meet this objection 
Eicuorst (11) (1875) completely severed the spinal cord of 
several dogs and was able to confirm his previous results. The 
article published by Masius (23) in 1880 describes the removal 
of a part of the sacral cord in young dogs. Microscopic ex- 
amination, made after healing had occurred, showed in the 
connective tissue many nerve fibers, mostly myelinated and 
grouped in bundles. These were continuous with the fiber 
tracts of the stumps. No nerve cells were found in the scar 
tissue. A month anda half after complete transection of the 
cord, STROEBE (30) (1894) found in the scar nerve fibers which 
had grown in from the dorsal roots. These fibers never com- 
pletely crossed the line of the incision. He noticeda few 
mitotic figures in the ependyma cells lining the central canal, 
but could not show that they bore any relation to the scar 
formation. Barr, Dawson and Marsnatt (1) (1899) report 
some careful experiments on dogs, which led them to believe, 
on the basis of physiological evidence, that the dorsal root 
fibers, which they had previously destroyed, had re-established 


Ranson, fibers in Lesions of the Brain. 189 


connections within the cord; but no microscopic examination 
was made. 

A case of injury to the human spinal cord asa result of 
spinal fracture in the thoracic region, has been described by 
Borst and Nicorarer (3) (1897). Here the scar contained 
numerous nerve fibers which were considered as ingrowths from 
peripheral nerves, that is, of course, from the dorsal roots. 

It is worth noticing that fibers in the cicatrix are described 
by Eicuorst, STROEBE, and Borst and NIcOLAIEgR, as arising 
from the spinal ganglia, or, what amounts to the same thing, 
from the dorsal roots of the spinal nerves. This may also be 
inferred from the results of BAER, Dawson and MarsHatt. 
Since it is known that the spinal ganglia always contain many 
undeveloped neurones (17), it is possible that these fibers are 
not regenerated fibers but represent neurones which have com- 
pleted their development under the stimulus of the injury. 
Kun (21) (1901) has made an observation quite analogous to 
this. He found that depriving an area of skin on a frog’s back 
of its sensory supply by sectioning one of the cutaneous 
nerves caused fibers to grow into that area from uninjured 
nerves. These, since they were not regenerating fibers, must 
represent axones which, being more or less incompletely devel- 
oped before the experiment, have been stimulated to complete 
their growth as a result of the injury. 

A large number of careful investigators have reported 
entirely negative results on the mammalian cord. ScHIEFFER- 
DECKER (28) (1876) practicing transection of the cord in dogs a 
few months old, was unable to demonstrate any nerve fibers in 
the resultant scar. WesrpeHaL (37) (1870) bored a small hole 
in the vertebral column and spinal cord but found no evidence 
of regeneration. KerkEszTszeGHyY (20) (1892), Scosso (29) 
(1891) and Marinesco (22) (1894) have also obtained negative 
results. 

It will be seen that the evidence does not encourage the 
belief that the fiber columns of the cord are capable of regen- 
ration, although fibers may grow into the scar from the dorsal 
roots. Masrus, however, found fibers which seemed to arise 


190 JOURNAL OF COMPARATIVE NEUROLOGY. 


from the severed tracts of the cord. It seems that nerve cells 
in the injured spinal cord of mammals do not proliferate. The 
ependyma cells lining the central canal may show indications of 
mitotic division (STROEBE); which, when taken in connection 
with the observations of BARFURTH and Caporaso on the regen- 
erative capacity of these cells in the lower vertebrates, is of con- 
siderable theoretical interest. 


Brain of Mammals. 


The healing of wounds in the brains of mammals has been 
studied by numerous investigators, with a view to determining 
whether new nerve tissue is formed. TEpEscHI (31) (1897) 
has given a good account of the reaction of brain tissue to 
various injuries. Wounds were inflicted by him on the 
encephala of rabbits, guinea-pigs, cats and dogs, some by 
plunging a hot needle into the tissue, some by the introduction 
of a foreign body (a piece of paraffin) and others by resection 
of a large piece of a hemisphere. He also studied the effect 
of subdural injection of pathogenic bacteria. Three days after 
the introduction of a foreign body he observed at about I mm. 
distance from ita zone of reaction. Some of the nerve cells in 
this region were altered in form and contained vacuoles, fat 
droplets, and darkly stained nuclei. Many other nerve cells 
were in various stages of karyokinetic division. In isolated 
nerve cells he observed an abnormal arrangement of the chro- 
matic substance of the nucleus, which might be interpreted as a 
phase in the dissolution of the cell. But there were also nor- 
mally constructed karyokinetic figures, which resulted in cell 
proliferation. Mitosis was also seen in the neuroglia and endo- 
thelial cells. Some months after the operation the foreign 
body was surrounded by tissue, consisting chiefly of neuroglia 
but also containing a few ganglion cells and many nerve fibers. 
This nervous tissue is new-formed, he argues, because it is 
found in the place of tissue which had suffered profound de- 
generative changes. He also pictures a nerve cell which has 
sent processes between the lamellae of paraffin and which must 
therefore have been developed since the operation. When re- 


Ranson, Fibers in Lesions of the Brain, IgI 


section of a part of a hemisphere was practiced, the place of 
the excised tissue was taken by a soft reddish mass composed 
of nerve fibers, nerve cells and neuroglia. These statements 
would be more convincing if we could be sure that no nerve 
cells had survived the degenerative processes in the immediate 
vicinity of the foreign body; and that when a piece of the 
brain was excised, the margins of the wound did not become 
approximated, at the same time undergoing partial degenera- 
tion with glia formation, and thus giving rise to the soft red- 
dish mass described as filling the wound. 

Analogous to TEDESCHI’s observations after excision of 
part of a dog’s cortex, are the results of Virzou (33) (1897) 
after ablation of the occipital lobes of a young monkey. 
About two years after the first operation he opened the skull 
again and in the second operation removed the tissue which 
was found occupying the position of the former occipital lobes. 
This mass, which he regarded as new-formed, was much the 
same in structure as normal brain tissue; but since it was re- 
moved under limitations imposed by the operation, neither its 
relation to the rest of the brain, nor its gross appearance was 
adequately described. Nor do his figures show the transition 
between the new-formed and the normal brain substance. For 
this reason one cannot be sure that it does not represent 
uninjured tissue displaced posteriorly. Whatever may be 
thought of the physiological evidence brought forward in this 
case, the anatomical evidence seems to be very inconclusive. 

With regard to the mitotic division of nerve cells, TE- 
DESCHI has received better support. FRIEDMANN (13) (1889) 
has observed that, when the irritation of the cortex is suffi- 
ciently severe and at the same time perfectly aseptic, prolifera- 
tion of the ganglion cells may occur. Monpino (25) (1886) 
and Coen (8) (1887) obtained similar results after puncturing 
the brains of guinea-pigs and rabbits with red hot needles. 
Coen, however, found that the new-formed ganglion cells soon 
disappeared, and the final cicatrix contained neither nerve 
fibers nor nerve cells. 

While most of the investigators who have worked in this 


192 JOURNAL OF CoMPARATIVE NEUROLOGY. 


field have observed mitotic figures in the nerve cells, not all 
agree that they result in cell proliferation. SANARELLI (27) 
(1891) used a hot needle on the cerebrum and cerebellum of 
rabbits. Three days later he found karyokinesis in the nerve 
cells, where it had not progressed in a normal manner, nor pre- 
sented all the stages of true mitosis. At this time there 
appeared in the normal brain substance, elements resembling 
giant nuclei which he believes are the results of incomplete, 
atypical karyokinesis of the nerve cells. After eighty days the 
stab canal was filled with pure connective tissue, which was sur- 
rounded by a zone of neuroglia, and this in turn bordered on 
the normal brain substance. Nerve cells were not present in 
the connective tissue nor in the neuroglia formation. 

MarInEsco (22) (1894) using the method of Corn and 
SANARELLI on the brains of young rabbits, guinea-pigs and 
cats, observed in the ganglion cells karyokinetic changes which 
did not, however, reach the ‘‘meta”’ stage. The cicatrix which 
formed later contained neither nerve fibers nor nerve cells. 
TscHISTOWITSCH (32) (1898) found that brain wounds in young 
rabbits and dogs healed with pure connective tissue. 

It is plain that nerve cells may begin to proliferate, some- 
times not reaching the stage of metakinesis (MARINESCO), 
sometimes producing giant nuclei (SANARELLI), and at other 
times proceeding to cell division, with rapid disappearance of 
the new-formed cells (CoEN). It is doubtful whether these new 
cells ever continue to exist, although they are reported to do 
so by Tepescur. It will be seen that observations have been 
confined largely to the nerve cells, the fibers receiving little 
attention. 

Several investigations have been made upon scars from 
lesions in the human brain. WorcesTER (38) (1898) describes 
a case of complete degeneration of the right half of the cor- 
pora quadrigemina in a woman of 51 years. The region is 
filled with neuroglia. There are no nerve cells; but in the 
midst of the area is a much convoluted bundle of nerve fibers 
which appears to have grown in from the region of the red nu- 
cleus. ZIEGLER (39) (1876), KaLpEn (19) (1891), and CHEN- 


Ranson, Fibers in Lesions of the Brain. 193 


ZINSKI (7) (1902) have failed to find any nerve fibers or nerve 
cells in the scars resulting from wounds in the human brain. 


IT, Experimental Observations. 


Introduction. 


We know that not all the neurones are fully developed at 
birth. The number of undeveloped neurones in the young rat 
is especially large. Dr. Axtice Hamitron has shown that 
mitotic figures are very abundant in the developing central 
nervous system of the white rat four days old(14). Harari has 
shown that the number of nerve fibers in the dorsal roots of 
the spinal nerves of the rat increases steadily long after the 
animal is mature (17). The increase in the number of fibers 
in the dorsal roots he explains as due to the development of 
the small, non-functional cells of the spinal ganglion into large 
functional cells. He has been able to show that this transfor- 
mation does occur (17). He has recently made similar obser- 
vations on the ventral roots (18), showing also in accordance 
with the observations of Harpesty (15) on the frog, that the 
number of fibers in the ventral root is greater in sections near 
the cord than in those at some distance from it. These 
observations indicate the presence in the ventral roots of nerve 
fibers that have not completed their growth. If we may infer 
the same method of growth in the central nervous system— 
and Dr. Hamitton’s (14) observations would indicate that such 
an inference is justifiable—we may assume that new nerve 
fibers are constantly being formed in the brains of young rats 
and probably in the brains of other young mammals. The 
failure to find nerve fibers in the scar when young animals have 
been operated on involves, therefore, the conclusion that nerve 
fibers which would normally have developed through this 
region have failed to penetrate the cicatricial tissue. While it 
is not inconceivable that the dense fibrous tissue filling the 
wound should act as a barrier, it seemed probable that if 
sufficiently immature animals were used nerve fibers would be 
found crossing the site of the lesion. At the suggestion of 


194 JOURNAL OF COMPARATIVE NEUROLOGY. 


Professor DonaLpson, I undertook a series of experiments to 
test this hypothesis. 


Technique. 


For the purpose of this investigation young rats were 
especially suited, because of their immature condition at birth. 
According to a recent observation by Watson (36) (1903) in 
this laboratory, the nervous system of the new born white rat 
does not contain a single medullated nerve fiber. During the 
first few weeks the increase in the weight of the brain is very 
rapid. For example, calculating froma series of records on 
the brain weight of the white rat collected by Professor Don- 
ALDSON, the youngest rat used in the investigation gained 500% 
in brain weight (from .2 g. to 1.0 g.) during the forty days fol- 
lowing the operation. This increased weight is due in large 
part to the formation of new nerve fibers. We should expect 
that tissue undergoing such rapid development would react to 
an injury quite differently from that more nearly adult. Pre- 
vious investigations have, however, been made largely on adult 
animals, or on those in a much later phase of development. 
Those used for this study were at the time of the operation 
aged respectively, No. 1, 21 days; No. 2,7 days; \No:iaye 
days; No. 4, 0.5 days (12 hours). 

The operation was very simple. A sharp-pointed thin 
bladed scalpel, 1 mm. in breadth, was passed through the soft 
skull and run 2 or 3 millimeters into the brain substance, far 
enough to cut through the corpus callosum. The wound was 
made in each case parallel to, and about 1 mm. to the left of 
the great longitudinal sinus and 2 mm. in front of the lateral 
sinus. These sinuses, which mark the medial and posterior 
boundaries of the cerebral hemispheres, are clearly visible 
through the membranous skull. They served as guides in 
making the operation, enabling one to cut into very nearly the 
same region in each brain. The operation consisted of a single 
stab, so producing a wound the shape and size of the end of 
the knife. 

The usual aseptic precautions were observed; the instru- 


Ranson, Fibers in Leszons of the Brain. 195 


ments were sterilized in boiling water; the site of the wound 
was freed from hair and washed in mercuric chloride, I-1000. 
After the operation the opening was sealed with collodion. 

It was not until many failures had been made that success 
was attained in raising the rats operated on during the first 
week of life. In the first cases when they were returned to the 
nest the mother ate them, and when fed by hand they could be 
kept alive for only a short time. Others working in this labo- 
ratory have experienced the same difficulty. It was supposed 
that the action of the mother was caused by some foreign odor 
clinging to the young rat after it had been handled. A certain 
degree of success, however, was secured by keeping the young 
in an incubator heated nearly to the body temperature, and 
forcing the mother, kept in a cage near by, to nurse them at 
regular intervals (every four hours during the day and eight 
hours at night). By this means some young were kept alive 
for ten days. It was found that as long as the young were 
kept warm the mother could safely be left with them and that 
after a time she would care for them of her own accord. It 
was also found that if the animals were taken from the nest 
under conditions which prevented the loss of body-heat, they 
could be returned to the mother with safety. For this purpose 
a jar was warmed to body temperature and filled with hot cot- 
ton. In this the rat was carried from the nest to the opera- 
ting room, where it was placed in an incubator until all was 
ready. The rat was held ina hot cloth during the operation, 
which was performed in a few minutes and without anaesthet- 
ics. It was then returned to the hot jar and carried back to 
the animal room. When these precautions were observed the 
mother never refused to care for the young. 

After the operation the rats were allowed to live for one 
month anda half. The brains were then removed, hardened 
in MULLER’s fluid at 40° C. for one month and imbedded in cel- 
loidin. The occipital region of each brain was cut in serial 
frontal sections 45 yw thick and stained by the PaLt-WEIGERT 
technique. Also for comparison the brain of a rat, operated 
on at the age of four days and killed thirty-two days later, was 


196 JOURNAL OF COMPARATIVE NEUROLOGY. 


hardened in vAN GEHUCHTEN’S solution, imbedded in paraffin, 
cut in serial sections 6 » thick, and stained in erythrosin and 
toluidin blue. 


Results. 


It will be most convenient to give a general description of 
the wounds in each of the four animals, and to follow this with 
a summary of the more important points as they are seen when 
the four specimens are compared. It will be found that this . 
description differs from the usual accounts in the following 
points: (1) the absence of adhesions in the meninges, (2) the 
absence of a well defined connective tissue scar, (3) the 
atrophy of sectioned nerve fibers on the cell body side of the 
injury, (4) the distortion of the wound due to the shifting of 
areas in the growing cortex and (5) the presence of nerve fibers 
crossing the site of the lesion. 

Rat No. I was operated on when twenty-one days old and 
was killed fifty-three days later. The wound was found in 
the occipital region of the left hemisphere, parallel to, and near 
the middle line. On examining the serial sections of this 
region the wound could be traced from near the posterior edge 
of the splenium forward for a little over 2 mm. __In its anterior 
part it penetrates the cornu Ammonis and runs a slight distance 
into the thalamus. From this point the floor of the wound 
gradually slopes upward and backward until the surface of the 
cortex is reached. By reference to Fig. 1 the general charac- 
ter of the wound will be seen. It is visible asa light band 
running down from the surface of the cortex and extending 
just through the corpus callosum, but it does not at this point 
extend into the cornu Ammonis. Throughout its whole extent 
dark granular masses are apparent—the remains of unabsorbed 
blood. The scar tissue is quite abundant in this brain clearly 
marking off the wound from the normal brain substance. On 
comparing this preparation with a corresponding one ina set of 
serial sections of a normal rat brain, it was found that the bank- 
ing up of fibers on the medial side of the wound is not entirely 
the result of the operation, as an evident excess of fibers on the 


Ranson, /2bers in Lestons of the Brain. 197 


medial side is to be seen at a corresponding level in the normal 
brain. It happened that the knife passed along the line that 
forms the lateral limit of the medial area of dense fibers, and 
the presence of a band of scar tissue almost devoid of fibers 
enhances the contrast to some extent. 

The light band which marks the path of the knife is al- 
most devoid of fibers; but a few isolated ones may be dis- 
tinguished in the scar. Although clearly visible in the original 
preparation, they do not appear in the photograph which was 
taken with a very low power. In order to show these more 
accurately than could be done by a drawing on a small scale, a 
camera lucida tracing has been made (Fig. 5.). This tracing 
was taken from the same preparation that was used for the 
photograph, and represents a point in the lesion one-third of 
the way from the corpus callosum to the surface of the cortex. 
This preparation is No. 27 inaseries of 45 passing through this 
wound; it is therefore at some distance from either the anterior 
or posterior extremity of the lesion. These new formed fibers 
follow a very irregular course, so that it is not possible to trace 
them far in the plane of any one section. But in a number of 
cases it is possible to follow single fibers from the normal brain 
substance on one side through the scar tissue and into the nor- 
mal brain substance on the other side. Such a fiber is seen in 
oes 

Thus two tests were rigorously applied before the conclu- 
sion was formed that the fibers in the scar were really new 
elements; Ist, The fiber in question must pass completely 
across the lesion, making connections with the brain tissue on 
both sides; and 2nd, The fiber must cross near the geometric 
center of the lesion. Fibers that passed into the scar and then 
back again into the brain substance on the side from which 
they emerged were not counted as new-formed, for these might 
have been displaced into the opening left by the’ knife. One 
could not be sure that they had been formed since the opera- 
tion. Fibers that passed across the lesion near its anterior or 
posterior extremities or near its lower limit were also disre- 
garded; for these might have been uninjured fibers crowded 


198 JOURNAL OF COMPARATIVE NEUROLOGY. 


into the degenerated areas. But a fiber such as is illustrated in 
the tracing, crossing the wound near its center and making 
connections with the brain tissue on both sides of the scar, 
must without doubt be new-formed. 

Rat No. 2 was operated on when seven days old and 
killed forty-two days later. The wound in the skull was indi- 
cated only by a slight roughness, ossification at this point being 
as complete as elsewhere. The dura mater was normal and 
without adhesions to the scar in the skull or brain. The brain 
scar is not visible, its locality being indicated only by the 
arrangement of the surface veins which encircle it. 

In serial sections the lesion is clearly defined for a distance 
of 1.5 mm., of which 1.2 mm. lies posterior to the line of the 
splenium. Throughout its whole extent the wound passes 
through the substantia alba of the occipital lobe, into the cortex 
covering its under surface. Scattered along through the white 
matter there is a considerable amount of unabsorbed blood pig- 
ment which was, however, largely absent from the section of 
which Fig. 2 isa photograph. As will be seen by reference to 
that figure, the cortex on the upper surface of the hemisphere 
has healed so completely that it is hardly possible to make out 
the path of the knife in the upper two-thirds of the grey mat- 
ter, while in its lower third and in the substantia alba the lesion 
is clearly visible. Although the distribution of the fibers in 
the upper region of the cortex is quite normal, the knife must 
have passed through this area to have reached the structures 
below. 

Somewhat lateralward to the wound in the upper part of 
the hemisphere is a peculiar aggregation of fibers in the cortex 
covering the ventral surface. This region is shown in more 
detail in Fig. 9. The dark granular masses indicate the line 
along which the knife has passed. Numerous fibers running 
downward from the white matter are seen to follow this line 
and on approaching the ventral surface to spread out in a fan- 
shaped area. Nerve fibers in considerable numbers may be 
seen crossing the site of the lesion. Especially is this notice- 
able just below the substantia alba (Ff). 


Ranson, Fibers in Lesions of the Brain. 199 


It will be seen (Fig. 2) that the lines indicating the path of 
the knife above and below the white matter are not continuous, 
that below being more lateralward than that above, and running 
at a different angle. Since the injury was produced by a straight 
stab, this appearance can be explained only on the supposition 
that the areas of the cortex have shifted their relative positions 
during the process of growth. The shifting that takes place 
in the cortex over the corpus callosum is just the reverse, as 
seen in the sections near the anterior extremity of this same 
wound, where the part above the corpus callosum is more 
lateralward than the part below. 

Rat No. 3 was operated on when three days of age and 
killed forty-one days later. No sign of the injury could be 
found in the skull or the dura mater. On the brain a pale pink- 
ish groove, I.5 mm. in length, marked the line along which 
the knife had entered the cortex. 

When serial sections are examined, it is observed that the 
healing process is more complete than in the brains of the two 
older rats. The remains of unabsorbed blood are scarcely to 
be found; and at its anterior and posterior extremities the 
wound fades off so gradually into the normal brain substance 
that it is impossible by a study of serial sections to determine 
its exact limits. Its posterior extremity is slightly behind the 
end of the splenium, from which point it runs forward fora 
distance of a little more than 1.5 mm. _ By reference to Fig. 3 
the general appearance of the wound will be seen. _Lateral to 
the point of the lesion the corpus callosum has entirely disap- 
peared from the injured region. A small portion of its fibers 
remains intact near the middle line; but elsewhere the sub- 
stantia alba of the cortex rests immediately on the cornu 
Ammonis and the alveus. The alveus is absent for a short 
distance. The point near the middle line where the corpus 
callosum and alveus end represents the original position of the 
wound. By the large growth of fibers in the cortex medial to 
the lesion, the scar has been shifted lateralward until it has 
assumed the shape of a dilated V. 

The disappearance of the corpus callosum lateralward to 


200 JOURNAL OF COMPARATIVE NEUROLOGY. 


the injury is of special interest. Had the sectioned fibers of 
the callosum merely degenerated in the part distal to the 
lesion, there would have been a considerable number of fibers 
in the corpus callosum on each side of the injury, instead of an 
entire absence of them on the lateral side. Their absence 
on one side and presence on the other can be under- 
stood by supposing that the fibers cut near their cells of 
origin—fibers whose cells were situated in the injured hemis- 
phere—degenerated throughout their whole extent, while 
fibers cut at a greater distance from their cells of origin— 
fibers whose cells were situated in the opposite hemisphere— 
degenerated upon the distal side of the lesion only. In von 
GuDDEN’s Gesammelte Abhandlungen (34) there is recorded a - 
similar observation. Six or eight weeks after removal of one 
cerebral hemisphere in a young rabbit, he found that the ante- 
rior commissure and corpus callosum had completely atrophied 
in the remaining hemisphere. 

In the upper region of the cortex the wound is recognized 
with difficulty, because there is so little scar tissue and the 
number of crossing fibers is so great. It can, however, be 
seen as a faint line along the path the knife must have taken to. 
reach the underlying parts where the injury is more apparent. 
Fig. 6 is a camera lucida tracing of some of the fibers in the 
scar. It was taken from the same preparation as Fig. 3 about 
one-third the way up from the corpus callosum to the cortex. 
Two fibers fulfil the requirement of passing into the brain 
tissue on each side, and the others would probably do so if 
they could be followed far enough. 

Rat No. 4 was twelve hours old at the time of the opera- 
tion and forty days old when killed. The meninges were quite 
normal, and there was no sign of the injury apparent on the 
surface of the brain. Serial sections revealed a lesion extend- 
ing 1 mm. behind the posterior extremity of the splenium and 
almost 14 mm. in front of that point. The general appearance 
of the wound is shown in Fig. 4. The upper part of the 
cortex is normal, no scar tissue can be seen there, and nerve 
fibers have a normal distribution. In the lower half is a trian- 


Ranson, F2bers in Lesions of the Brain. 201 


gular area lighter than the rest, representing scar tissue that 
has taken the place of a large blood clot. This runs through 
all the sections, giving, when reconstructed, a mass in the 
shape of a three-sided prism with base directed ventrally. 
Below this lies the corpus callosum degenerated on both sides 
of the lesion. In this case the axons degenerated toward the 
cell whether cut before or after crossing the middle line. <A 
few straggling fibers are seen in the otherwise unstained area. 
These probably represent neurones which have completed their 
development since the operation. 

In the tissue composing the scar numerous nerve fibers 
pass in every direction interlacing to form a large meshed 
irregular network (Fig. 7). Fig. 8 is a camera lucida tracing 
of a part of this same area from another section of this brain. 
Owing to the unusual width of the scar in this region only a 
part of it could be represented in the tracing; the brain tissue 
along its margins is not indicated as in previous tracings. 

The presence of medullated nerve fibers in the cicatricial 
tissue, demonstrated in each of the four brains, might.be ex- 
plained on either of two hypotheses. They might represent 
regenerated portions of nerve fibers cut by the operation ; or 
they might have been formed since the operation by outgrowths 
from cell bodies incompletely developed at that time. Al- 
though there is nothing in the appearance of any given section 
to exclude the former theory, the latter seems the more prob- 
able. This conclusion is supported by the uniformly negative 
results obtained by those who have studied wounds in the brains 
of adult mammals; and by the fact that in the four cases 
given here the number of fibers in the scar diminishes regularly 
as the age of the animal at the time of the operation increases. 
(Compare Fig. 8 with Figs. 6 and 5). These facts might be 
explained as indicating a loss of the regenerative capacity of 
the neurones within the brain as the animal becomes older; but 
are more probably due to the diminution in the number of un- 
developed cells capable of sending axons across the lesion 
subsequently to the injury. During the period of active growth 
with which we have to deal the diminution in the number of 


202 JoURNAL OF COMPARATIVE NEUROLOGY. 


these cells would be sufficiently rapid to explain the decrease 
in the number of the crossing fibers. An observation, which 
tells strongly against the regeneration theory in this case, is. 
that the fibers of the corpus callosum degenerated toward their 
cells of origin after being cut in the two younger rats, while 
they remained intact on the cell side of the injury in older rats. 
This observation shows that nerve fibers instead of possessing 
unusual regenerative power in these young animals (where the 
largest number of nerve fibers were found in the scar) tend to 
degenerate completely after injury. 


Summary. 


(1). The adhesions which are described by other investi- 
gators as binding the brain scar to the meninges and the men- 
inges to the tissue filling the skull wound were entirely absent 
in these young rats. 

(2). Very little scar tissue was found in the brain. In 
the upper part of the cortex it was not noticeable in Pat- 
WEIGERT preparations; in the substantia alba it was more abun- 
dant. With the exception of the large triangular area in the 
brain of rat No. 4, there is a steady decrease in the amount of 
scar tissue with the decrease in the age of the animal at the 
time of the operation. How little connective tissue there is 
in the wound is seen in Fig. 10. Nerve cells quite normal in 
appearance border directly on the line that represents the path 
of the knife. Only here and there cana connective tissue 
cell be found in the scar (c). Inthe brains of older animals 
the connective tissue is very abundant at the site of the lesion. 
It may be that the absence of a dense connective tissue scar in 
the cases here given may in part explain the positive results 
obtained on the crossing of nerve fibers, but the increase in the 
number of nerve fibers is out of proportion to the decrease in 
the amount of scar tissue. 

(3). A degeneration of the fibers of the corpus callo- 
sum toward their cells of origin was noticed in the two younger 
rats. 

(4). Shifting of the parts of the cortex with reference to 


Ranson, F2bers in Leszons of the Brain. 203 


each other occurred in all but the oldest rat. In rats Nos. 2, 
3 and 4 it was found that the parts of the cortex had altered 
their relative positions in the same manner, such that if we 
consider the under surface of the cortex as stationery the cor- 
tex upon the upper surface shifts lateralward over the splenium, 
Fig. 3, and medialward in the posterior part of the occipital 
lobe, Fig. 2. This is explained by the fact that when a large 
number of fibers develops in any area, that area increases in 
size more rapidly than the surrounding parts. in Fig. 3 is seen 
a dense area of fibers just medial to the scar, which is respon- 
sible for the lateral displacement of the scar. 

(5). We have seen that atthe center of stab wounds, 
where every fiber must have been cut by the knife, medullated 
nerve fibers may be traced across the wound and into the norm- 
al brain tissue on each side. While there can be no doubt 
that these are new formed elements, they represent in all prob- 
ability not vegenerated but entirely new axons. The number 
of such fibers is very great in the youngest rats, but decreases 
quite rapidly as the age of the rat at the time of the operation 
increases. Unfortunately the brains of rats operated on ata 
more advanced age were spoiled in the process of hardening, 
and I cannot say at what time the brain of the rat ceases to 
have the capacity of sending fibers across the site of the lesion. 


BIBLIOGRAPHY. 
Baer, W. S., Dawson, P. M. and Marshall, H. T. 


1. ’99. Kegeneration of the dorsal root fibers of the second cervical 
nerve within the spinal cord. /. Axper. A/., IV, 29. 
Barfurth, D 


2. ‘91. Zur Kegeneration der Gewebe. arch. f. mikr. Anat. 
XXXVII, 406. 
Borst, M. 
3. °97. Regenerationsvorginge im Riickenmark nach Verletzungen. 
Verhanat. ad. phjs.-med. Gesellsch. zu Wiirzb. N. ¥F., XXXI, 270. 
Brown-Séquard, M. 
4. ’51. Gaz. med. de Paris, 1851, p. 447. 
5. °’92. De la régénération de la moelle épiniére d’aprés l’expérimenta- 
tion et des faits cliniques. Arch. de Phys. S. 5, T. IV, 410. 


204 JOURNAL OF COMPARATIVE NEUROLOGY. 


Caporaso, L. 
6. ’87. Sulla rigenerazione del midollo spinale della coda di tritoni. 
Ziegler’s Bettrige. WV, 67. 
Chenzinski, C. 
7. ’o2. Zur Frage iiber die Heilung der Hirnwunden. Centralblatt f. 
allg. Path. u. path. Anat, XIII, 161. 
Coen, E. 
8. °’87. Ueber die Heilung von Stichwunden des Gehirns. Beztrage 
z, path. Anat. u. Phy, II, 107. 
Danielewsky, B. 
9. ’91. Ueber die Regeneration Grosshirnhemispharen beim Frosch. 
Verhand. des X. Internat. Med. Cong. I, 18, Berlin, 1891. 
Eichorst und Naunyn. 
10. °74. Ueber die Regeneration und Verinderungen im Riickenmark 
nach streckenweiser totaler Zerstérung desselben. Arch. fiir exp. 
Path.u. Phar. I (quoted by MAstIus). 
Eichorst, H. 
11. °75. Ueber Regeneration und Degeneration des Riickenmarkes. 
Zeitschr. f. Klin. med., Berlin, 1, 284. 
Fraisse, P. 
12. ’Ss. Die Regeneration von Gewebe u. Organen bei den Wirbel- 
thiern. Kassel. uw. Berlin. (Quoted by TSCHISTOWITSCH). 
Friedmann, M. 
13. ’89. Zur Histologie und Formeneintheilung der acuten mit eitrigen 
genuinen Encephalitis. Mewrol. Centralblait, VIII, 441. 
Hamilton, Alice. 
14. ‘91. Division of differentiated cells in the central nervous sys- 
tem of the white rat. Journ. Comp. Neur. XI, 297. 
Hardesty, I. 
15. ’99. The number and arrangement of the fibers forming the spinal 
nerves of the frog (Rana virescens). Journ. Comp. Neur. IX, p. 65 
Harrison, R. G. 
16. ’98. Growth and regeneration of the tail of the frog larva. 
Roux. Arch. VII, 430. 
Hatai, S. 
17. ’o2. Number and size of the spinal ganglion cells and dorsal root 
fibers in the white rat at different ages. Journ. Comp. Neur., XII, 107. 
18. °03. Onthe increase in the number of medullated nerve fibers in 
the ventral roots of the spinal nerves of the growing white rat. 
Journ. Comp. Neur. XIII, No. 3. 
Kalden. 
19. ’91. Ueber die Heilung von Gehirnwunden. Centralblatt f. path. 
Anat. 1891. (Quoted by STROEBE). 
Keresztszeghy, J. 
20. °92. Ueber Degenerations- und Regenerations-vorginge am Riick- 
enmarke. Ziegler’s Bettrége, X11, 33. 


Ranson, Fibers zn Lesions of the Brain. 205 


Kiihn, A. 

21. ’o1. Weiterer Beitrag zur Kenntniss des Nervenverlaufs in der 

Riickenhaut von Rana fusca. Arch. f. mikr. Anat. LVII, 445. 
Marinesco, G. 
22. ’94. Sur la régénération des centres nerveaux. Compt, rend. Soc. 
de Biol. i0s., I, 389. 
Masius, M. 
23. 80. De la régénération de la moelle épinére. Arch. de Biol. 1880. 
Masius, M. und Vanlair. 

24. ’69. Anatomische und functionelle Wiederherstellung des Riick- 

enmarks beim Frosch. Centralblatt f. d. med. Wissensch. 1869. 
Mondino, C. 

25. 86. Sulla cariocinesi delle cellule del Purkinje consecutiva ad 
irritazione cerebellare. Archivio di Scienze Penali ed Anthropologia 
Criminale, 1886. (Quoted by CoEn.) 

Miller, H. 

26. ‘64. Ueber Regeneration d. Wirbelsdule und d. Riickenmarkes 

bei Tritonen und Eidechsen. Frankfurt. 
Sanarelli, G. 

27. ’91. Reparationsvorginge im Gross- u. Kleinhirn. Centralblatt f. 

allgem. Path. I1, 429. 
Schiefferdecker, P. 

28. ’76. Ueber Regeneration, Degeneration und Architectur des 

Rittckenmarkes. Virchow’s Arch. LXVII, 542. 
Sgobbo, F. 

29. ’gt. Sulla rigenerazione del midollo spinale nei vertebrati. Za 

Psichiatria, VIII. (Quoted by GLEY.) 
Stroebe, H. 

30. ’94. Experimentelle Untersuchungen iiber die Degeneration und 
reparatorischen Vorgange bei der Heilung von Verletzungen des 
Riickenmarks. Zzegler’s Bettraége, XV, 383. 

Tedeschi, A. 

31. ’97. Anatomisch-experimenteller Beitrag zum Studium der Regen- 
eration des Gewebes des Centralnervensystems. Zzegler’s Beitrage. 
MOG Age 

Tschistowitsch, T. 

32. ’98. Ueber die Heilung aseptischer traumatischer Gehirnverlet- 

zungen. Ziegler’s Beitrége, XXIII, 321. 
Vitzou, A. 

33. ’97. La néoformation des cellules nerveuses dans le cerveau du 

singe. Arch. d. Phys. norm. et path. IX, 29. 
von Gudden, B. 

34. ’89. Gesammelte und Hinterlassene Abhandlungen. p. 130. 
Wirzburg, 1889. 

von Voit. 

35. ’68. Beobachtungen nach Abtragung der Hemispharen d. Gross« 
hirns bei Tauben. Sitzungsberichte d. Konigl. Bayr. Akad. d, Wise 
senschaft. 11,105. Miinchen, 1868. (Quoted by CoEn.) 


206 JOURNAL OF COMPARATIVE NEUROLOGY. 


Watson, John B. 
36. ’03. Animal education: an experimental study on the psychical 
development of the white rat, correlated with the growth of its 
nervous system. Zhe University of Chicago Press, Chicago, 1903. 
Westphal. 
37. ’70. Ueber kiinstlich erzeugte secundadre Degeneration einzelner 
Riickenmarksstringe. Arch. f. Psychiatrie u. Nervenkrankhetten, 
II, 415. 
Worcester, W. L. 
38. ’98. Regeneration of nerve fibers in the central nervous system. 
J: of, Exp. Meds Nis. 
Ziegler, E. 
39. ’95. Lehrbuch der patholog. Anat., Theil II, S. 354. 


EXPLANATION OF FIGURES. 
PLATE VIi. 


Figs. 1-4. Photo-micrographs of frontal sections of the left cerebral hem- 
ispheres of rats, showing the position of the scar and the general appearance 
of the surrounding parts. Sections 45 stained by the PAL-WEIGERT technique. 
The cut is indicated by ‘‘c’’. The scar is seen as a light band, medial to which 
the fibers are numerous, and lateral to which they are less abundant. 


Fig. 7. From Rat No. 1, operatedon when twenty-one days old and killed 
53 days later. 

Fig. 2. From Rat No. 2, operated on when 7 days old and killed 42 days 
later. Showing aggregation of fibers in the scar in the ventral part of the cor- 
tex just above the sharp notch on the ventral surface. Notice that the sear has 
been distorted by the shifting of the areas of the cortex with reference to 
each other. 

Fig. 3. From Rat No. 3, operated on when 3 days old and killed 41 days 
later. Notice the complete disappearance of the corpus callosum lateral to 
the injury. 

Fig, 4. From Rat No. 4, operated on when 0.5 day old andkilled 40 days 
later. Notice the degeneration of the corpus callosum on both sides of the 
lesion. 

Fig. 5. Camera lucida tracing of the fibers in and about a small area of 
the scar (Rat No. 1) seen in Fig. 1, taken at a point about one-third of the way 
from the corpus callosum to the surface of the cortex. The arrow indicates 
the path of the knife, and ‘‘f”’ a fiber crossing the scar. 

Fig. 6. Camera lucida tracing of the fibers in and about a small area of 
the scar (in Rat No. 3) seen in Fig. 3, taken at a point about one-third of the 
way from the corpus callosum to the surface of the cortex. The arrowand ‘‘e”” 
have the same significance as in the preceding figure. The fibers in the scar 
are slightly more abundant than in Fig. s. 

Fig. 7. Drawing of the area of scar tissue (Rat No. 4) seen in Fig. 4, 


“THE JOURNAL OF COMPARATIVE NEUROLOGY. VoL. XIII. 


ce 
' 
2 ' 
' 
ee] 
I 
' 
{ 
b 


Fig. J Fig. 3 i; 
Fig. 9 


Fig. 2 


> 


PLATE Vil. 


Ranson, f2bers in Lesions of the Brain. 207 


made onasmall scale because the scar was too wide to be included with 
the surrounding tissue in a high power tracing. The alveus (a) is normal. The 
corpus callosum (¢.c.) is unstained, with the exception of a few fibers. The 
arrow indicates the path of the knife and ‘‘f”’ designates a fiber apparently 
crossing the scar. 

fig. 8. Camera lucida tracing of a small area of scar in another section 
from the same brain (Rat No. 4) as that represented In Fig. 4. 

fig. 9. Drawing of the aggregation of fibers in the scar in the ventral 
part of the cortex (Rat No. 2) shown in Fig. 2. Notice the masses of blood 
pigment (4) along the line of the wound indicated by the arrow. Many fibers 
run downward along the scar, some of which cross the line of the incision at *f’’. 

fig. 10. Drawing from a frontal section of the left cerebral hemisphere 
of arat operated on at the age of 4 days and killed 32 days later. The cortex 
was fixed in VAN GEHUCHTEN’S solution, imbedded in paraffin, cut in sections 
6 thick, and stained in erythrosin and toluidin blue. The path of the knife 
is indicated by ‘‘c’’. Notice the absence of connective tissue cells, and the 
presence of nerve cells bordering directly on the line of the incision. 


ON THE DENSITY OF THE CUTANEOUS INNER- 
VATION IN MAN. 


By CuHaArRLes E. INGBERT. 
(Yrom the Neurological Laboratory of the University of Chicago.) 


I. INTRODUCTION. 


Il. THeE RELATION OF THE CUTANEOUS AND THE MUSCULAR NERVE 
FIBERS OF THE DORSAL ROOTS OF THE SPINAL NERVES OF MAN. 
rT. Lstimate Based on Stilling’s Results. 
Estimate Based on Votschvillo’s Results. 
3. Estimate Based on the Author’s Enumeration. 
4. Comparison and Discussion. 


ly 


III. DETERMINATION OF THE AREA OF THE DERMAL SURFACE OF THE 
HUMAN Bopy. 


I. Krause’s Determination. 
2. Funke’s Determination. 
3. Fubini and Roncht’s Determination. 
4. Mech’s Determination. 
5. Summary of Determinations. 
IV. THE INNERVATION OF THE DERMAL SURFACE OF THE HUMAN Bopy. 
1. Donaldson's Estimate. 
2. Votschvillo’s Estimate. 
3. Author's Estimate. 
4. Comparison and Discussion. 


V. THE BEARING OF THE AUTHOR’S ESTIMATE OF THE INNERVATION OF 
THE DERMAL SURFACE ON THE THEORY OF THE SPECIFIC ENERGIES 
oF NERVES. 


VI. SUMMARY. 
VII. BIBLIOGRAPHY, 


I. INTRODUCTION. 


It was, at first, the author’s intention to publish this dis- 
cussion as a part of his more extended paper on the ‘‘ Enumer- 
ation of the Medullated Nerve Fibers in the Dorsal Roots of 
the Spinal Nerves of Man.” This, however, was found im- 


210 JoURNAL OF COMPARATIVE NEUROLOGY. 


practicable owing to the accumulating evidence that STILLING’s 
estimate of the number of nerve fibers in the ventral roots of 
the spinal nerves, which was used in calculating the number of 
cutaneous fibers in the dorsal roots, was not sufficiently accu- 
rate for our purpose. The data necessary for this calculation 
are the enumerations of the nerve fibers in both the dorsal and 
the ventral roots. The former the author has published (1903), 
as already mentioned. It was therefore necessary to make a 
similar enumeration of the fibers in the ventral roots. This 
being now completed, the estimation of the density of the 
cutaneous innervation in man is possible and is here pre- 
sented. 


II. THE RELATION OF THE CUTANEOUS AND THE MUSCULAR 
NERVE FIBERS OF THE DorsSAL ROOTS OF THE SPINAL 
NERVES oF Man. 


Having determined the number of nerve fibers in the dor- 
sal roots of the left side in man (INGBERT, 1903), our next step 
is to inquire how many of these afferent fibers innervate mus- 
cles and other deep tissues, and how many the skin. 

According to SHERRINGTON’s (1894-95) observations on the 
cat, the afferent nerve fibers ina muscular nerve constitute two- 
fifths and the efferent three-fifths of its fibers. Or in other 
words, in a muscular nerve the afferent nerve fibers are to the 
efferent as 2 to 3. If we assume that this relation is typical of 
all muscular nerves, and if we assume that the same relation is 
true for man, then to the total number of efferent muscular 
nerve fibers, i. e., to the total number of nerve fibers in the 
ventral roots, there must be added two-thirds as many afferent 
fibers in order to make up the muscular nerves. In this calcu- 
lation we have neglected the recurrent nerve fibers in the ven- 
tral roots because, even if present in man, their number must 
be very small. The fibers of the ventral roots passing into the 
white ram communicantes have also been neglected, because 
their number is not great and because we may assume without 
any serious error that the relation between the afferent and 


INGBERT, Cutaneous Innervation in Man. 211 


efferent fibers in them does not differ sufficiently from that in 
other peripheral trunks to modify our result. 

I. Estimate Based on Stilling’s Results. According to 
STILLING (1859) there are in man, in round numbers, 500,000 
nerve fibers in the dorsal roots of the spinal nerves of both 
sides, and 300,000 in the ventral. We therefore take from 
500,000 fibers a number equal to two-thirds of 300,000, or 
200,000 fibers, in order to add these to the efferent fibers in the 
muscular nerves. This leaves 300,000 or 60% of the nerve 
fibers of the dorsal roots to innervate the dermal surface of the 
body. If these data were correct, as unfortunately they are 
not, we could conclude that about 60% of the afferent fibers in 
the dorsal roots of man are cutaneous, i. e., go to innervate the 
dermal surface of the body, and about 40% go to the muscles. 
In the 60% credited to the skin there are included the fibers 
which pass to the viscera by the vam communicantes and recur- 
rent nerves, if any, but so far as we know, the number of these 
is small, and, for the present purpose, may be neglected. 

2. Estimate Based on Voischvillo’s Results. Another 
calculation on the relation of the cutaneous and the muscular 
nerve fibers in the dorsal roots of man can be made from the 
results of VoIscHvILLo (1883) and those by myself. Vorscu- 
VILLO sectioned the peripheral nerves in man at the places 
where they divide into cutaneous and muscular branches. The 
material used by him was hardened by 1% osmic acid and pre- 
served in g5% alcohol. The areas of the cross-sections of the 
nerves (of which I can find no record) were obtained by two 
methods, (1) by dividing the volume (determined by weighing) 
of a piece of nerve by its length, (2) by projections of the 
cross-section made by means of a camera lucida. 

In estimating the number of fibers in a peripheral nerve 
VoIscHVILLO counted all the nerve fibers of the section of the 
nerve that could be seen within a certain number of the squares 
of the ocular micrometer, and from this result he estimated the 
number in the entire cross-section. 

VOISCHVILLO’S average for the number of nerve fibers in 
the cutaneous nerves derived from the brachial plexus of one 


212 JOURNAL OF COMPARATIVE NEUROLOGY. 


side is 119,337. Since, according to my count, the dorsal 
roots (C. V.—Th. I) which help form the brachial plexus, con- 
tain 193,095 fibers, the cutaneous fibers in these roots as 
determined by VoIscHvILLO amount to 61.7%. In consider- 
ing the dorsal roots C. V.—Th. I as the roots giving rise to 
the afferent fibers of the brachial plexus I have made use of P. 
EIsLer’s drawings (Rauber 1893). According to these draw- 
ings the brachial plexus also receives a branch from the roots of 
C. IV and another from the root Th. II (N. intercostobrach- 
jalis). These two sources of gain would make Volscu- 
VILLO’s results too large were it not for the fact that he omitted 
in his estimate the cutaneous fibers given off by root Th. I in 
the N. intercostalis primus, as well as the few cutaneous fibers 
in the rami dorsales of C. V to Th. I. These two factors no 
doubt counterbalance each other to a great extent and thus 
justify us in our calculation 

Again, by adding VolscHviILLo’s averages for the number 
of nerve fibers in the cutaneous nerves derived from the 
lumbosacral plexus of one side, I find them to amount to 154,- 
459 fibers (his own total, by another method which I cannot 
understand, is 82,167). According to my count, the dorsal 
roots (L. II—S. III) which help form this plexus contain 228,- 
117 nerve fibers. In other words, according to VoISCHVILLO’S 
results 67.7% of the nerve fibers in these dorsal roots are 
cutaneous. 

I have omitted the root L. I in this calculation because 
VoISCHVILLO made no estimate of the peripheral nerves de- 
rived from this root. The roots S. IV. S. V, and Coc. I, have 
also been omitted for the same reason. The gain from the root 
of L. I is, no doubt, counterbalanced to some extent by losses 
in branches passing into N. pudendus and N. clunium medius 
which he omitted. 

3. Estimate Based on the Author's Enumeration, Accord- 
ing to the author’s enumeration (INGBERT, 1903) the left dorsal 
roots of the spinal nerves of a large man contain 653,627 
medullated nerve fibers. Since the publication of this report 
the author has made an enumeration of the medullated nerve 


INGBERT, Cutaneous Innervation in Man. 2h3 


fibers in the ventral roots of the spinal nerves of the same cord. 
This latter enumeration gives 203,700 medullated nerve fibers 
on one, the left, side of the body, or about 407,400 on both 
sides. (The report on this investigation will be published in 
the near future. ) 

As already mentioned, according to SHERRINGTON 
(1894-95) there are two-thirds as many afferent nerve fibers in 
a muscular nerve (a mixed branch to a muscle) as there are 
efferent fibers. We must, therefore, take two-thirds of 203,- 
700 or 135,800 from 653,627 and add them to the efferent 
fibers in order to represent all the muscular nerves. This 
leaves 517,827 nerve fibers to innervate the dermal surface of 
the body; or, in other words, 79.22% of the fibers in the dor- 
sal roots to innervate the dermal surface and 20.78% to supply 
muscles and other deep tissues with afferent nerves. 

4. Comparison and Discussion. For comparison let us 
now repeat these estimates of the number of cutaneous fibers 
the in the dorsal roots: 


I. Estimate Based on STILLING’s Results 60.00% 
2. Estimate Based on VOISCHVILLO’s Results 
a. Brachial plexus 61.70% 
b. Lumbar plexus 67.70% 
3. Estimate Based on Author’s Results 79.22% 


The low figure obtained from STILLING’s data is doubtless 
due to SriLiine’s failure to include the fibers the diameter of 
which is less than 74, as the author has already demonstrated 
(INGBERT, 1903, p. 68), together with the fact that these fibers 
of asmall diameter are more abundant in the dorsal roots, a 
point to be discussed in another paper. We consequently con- 
clude that his figure for the number of fibers in the dorsal roots 
is more below the true one than his figure for those of the ven- 
tral roots. This becomes apparent on a comparison of the 
ratio between the fibers in the two roots. 


Ventral Roots. Dorsal Roots. 
STILLING a 1.66 
INGBERT I 3.20 


The cause of the low figure obtained from VoIscHVILLo’s 


214 JOURNAL OF COMPARATIVE NEUROLOGY. 


results is not so easily found; but is probably partly due to the 
fact that his results represent an estimate and not an enumera- 
tion. It is also highly probable that the osmic acid did not in 
the material used by him bring out all the smallest fibers suffi- 
ciently well to be included. | 

In order to determine the value of the calculations for the 
innervation of the skin about to be given, it will be well to 
consider in detail the methods by which the data for these 
calculations were obtained. 


III. DETERMINATION OF THE AREA OF THE DERMAL SuR- 
FACE OF THE Human Bopy. 


I. Krause’s Determination. Krause (1844) determined 
the dermal surface of the body to be 1,584,300 mm*. or about 
15 square feet. I can find no record of his method, nor any 
description of the subject used. : 

2. unke’s Determination. In making his determination 
of the area of the dermal surface FUNKE (1858) covered the 
half of a cadaver with gummed paper cut into one inch squares, 
as well as into smaller pieces, the values of which were deter- 
mined beforehand. In so doing care was exercised to cover 
every part of the skin, to allow no overlapping of the paper, 
and to guard against the folding of either the paper or the skin. 
The results obtained he considers accurate within a square inch. 
He thus determined the areas separately for each part of the 
body, and gives as the entire area 1,651,700 mm”. 

3. Fubinit and Ronchi’s Determination. These investiga- 
tors (1881) divided the surface of the body into its anatomical 
regions and marked these by sharp lines. These regions were 

‘then divided into such geometrical figures as could easily be 
measured. For the head a craniometer was employed. The 
measurements were made on the cadaver of a man 1.62 meters 
in height, and 50 kg. in weight. They determined the areas 
for the different parts separately and found the entire area to be 
1,606,685 mm’. 

4. Meeh’s Deiermination. Mee (1879) combined in his 
method several valuable features. "On large even surfaces he 


INGBERT, Cutaneous Innervation in Man. 215 


marked the regions by red lines and traced these on transparent 
paper. These areas were then measured from the paper. 
Around the fingers he wound strips of paper of a uniform 
width. Other uneven surfaces he covered with gummed paper. 
Although Meru measured several subjects, the results for a 
thirty-six years old man, corpulent, 171 cm. in height, and 
body-weight 78.2 kg., only are given. 
Area for the right side of the body: 


Head 80,380 mm?. 
Neck 45,060 ‘ 
Trunk 294,160 * 
Upper arm 785150) 
Fore-arm 67,860 ‘ 
Hand 535050) 
Thigh DOW 250) meine 
Shank 1265920) 5 
Foot 66,930 * 


Pelvic region 106,580 ‘* 


I.121,740 mm?. 
Both sides (his own figure) 2,243,490 mm?. 


By pelvic region MrEH means the dermal surface of the 
hips, i. e., the area between a line running along the crest of 
the pelvis and one transversely around the thigh about on a 
level with the ‘uberculum pubicum. ‘This region is innervated 
by lumbar and sacral nerves and is included in the area for the 
the entire leg. 

MeEeEn’s line between the neck region and the trunk region 
runs from the top of the acromion process along the clavicle on 
the ventral surface of the body and on the dorsal surface to 
about the level of the spine of the seventh cervical vertebra. 
Since, however, the nn. supraclaviculares innervate a portion of 
the skin below the clavicle, it is probable that his area for the 
neck region is a trifle too small for the skin innervated by the 
cutaneous nerve fibers in the dorsal spinal roots C. I-IV. 


216 JOURNAL OF COMPARATIVE NEUROLOGY. 


5. Summary of Determinations of the Area of the Entire 
Dermal Surface of the Human Body of Adult Men. 


Height Body wt. Area of 
imC€ms) ain) Ke: Dermal sur- 
Age face in mm?. Observer. 
17 years, well-built 169 55:5 1,920,550 MEEH (1879) 
20 years, 170 59-5 1,869,530 5G GC 
26 years, 162 62.3 1,895,960 66 oe 
36 years, corpulent 171 78.3 2,243,490 a ee 
36 years, emaciated 158 50.0 1,758,740 os ue 
45 years, 160 51.8 1,799,350 3 a 
66 years, 172 65.5 2,028,150 Ce uC 
(Not given) ? ? 1,584,300 KRAUSE (1844) 
(Not given) y ? 1,651,700 FUNKE (1858) 
(Not given) 162 50.0 1,606,685 FUBINI and BoONCHI (1881) 


Having now data for the number of cutaneous nerve 
fibers in the dorsal roots of the spinal nerves of man, and for 
the area of the dermal surface of the body, we are enabled to 
make calculations as to the innervation of the skin, not only 
for the body as a whole, but also for some of its principal parts. 


IV. THe INNERVATION OF THE DERMAL SURFACE OF THE 
Human Bopy. 


Z. Donaldson's Estimate. DoNALvson (1901) for the 
area of the dermal surface made use of the results obtained by 
Meen for a man weighing 136 lbs.—1,g00,000 mm’. For the 
number of cutaneous nerve fibers in the dorsal roots he took 
60% of STILL1NG’s estimate of 500,000, —300,000. According 
to this calculation, every cutaneous nerve fiber in the dorsal 
roots innervates, on the average, 6.3 mm” of the dermal 
surface. 

2. Vortschvillo’s Estimate. This investigator (1883) out- 
lined upon a cadaver the skin areas of the peripheral nerves as 
given by HENLE (1879). 

These outlines were traced on transparent paper, and the 
area of these outlines determined by placing this transparent 
paper upon a sheet of paper ruled in sq. cm. and sq. mm. 
According to his summary there are I19,337 cutaneous nerve 
fibers to innervate the 130,084 mm’. of dermal surface of the 
upper limb, or on the average one nerve fiber to every 
1.1 mm’. of skin; and 82,167 cutaneous nerve fibers (154,459 


INGBERT, Cutaneous Innervation in Man. 217 


according to my addition of his results) to innervate the 
303,566 mm?. of the dermal surface of the lower limb, or on 
the average one cutaneous nerve fiber for every 3.7 mm”. of 
skin. If my addition of his results be correct this will give 
one cutaneous nerve fiber to every I.g mm?. of skin. 

3. Author's Estimate. Since the cadaver from which the 
spinal roots used in my count of the nerve fibers was that of 
a man weighing 180 lbs. or 81.6 kg., I have made use of 
MEEn’s data, as already given, for the heaviest man measured 
by him, one weighing 78.3 kg. 

A. Body as a whole: 


Area of one side of body, 1,121,740 mm?. 
Deduction for skin of the head innervated by the 

cerebral nerves, (24 of 80,360) 53,573 mm?. 
Corrected area, 1,068,167 mm?. 
Number of cutaneous nerve fibers (79.22% of 653,627) 517,827 
1,068,167+517,827= 2.05 mm?, 


Hence, every cutaneous nerve fiber in the dorsal roots 
innervates on the average 2.05 mm”. of the dermal surface. 


Be) Armi: 
Area of the arm, 199,860 mm?. 
Number of cutaneous nerve fibers in dorsal roots, 
C. V.—Th. I (79.22% of 193,095) 152,970 
199,860+152,970—= 1.30 mm?, 


Hence, every cutaneous nerve fiber in the roots C. V— 
Th. I innervates, on the average, 1.30 mm? of the dermal 


surface of the arm. 
C. Head and Neck: 


Areaof one side of neck, 45,060 mm?. 
Area of one-third of head on one side, 26,786 ‘§ 
Total, 72,446 mm!. 
Number of cutaneous nerve fibers in dorsal roots of 
C. I.—IV (79.22% of 84,404) 66,865 
72,453+66,865= 1.08 mm?, 


Hence, every cutaneous nerve fiber in the dorsal roots of 
C. I—IV innervates, on the average, 1.08 mm’. of the dermal 
surface of the neck, and that part of the head not innervated 
by cerebral nerves. 


D. Leg: 
Area of surface of one leg, 501,680 mm?. 
Number of cutaneous nerve fibers in the dorsal roots 
L. I—Coc. I (79.22% of 258,502) 204,785 


501, 680+ 204,785= 2.45 mm?. 


218 JOURNAL OF COMPARATIVE NEUROLOGY. 


Hence, every cutaneous nerve fiber in the dorsal roots 
L. I—Coc. I innervates, on the average, 2.45 mm”. of the 
dermal surface of the leg. 


ES) Drunk: 
Area of surface, one side, 294,160 mm?. 
Number of cutaneous fibers in the dorsal roots of Th. 
II-XII (79-22% of 117,626) 93,182 
294, 160+-93,183= 3.15 mm?. 


Hence, every cutaneous nerve fiber in the dorsal roots 
Th. II-XII innervates, on the average, 3.15 mm”. of the 
dermal surface of the trunk. 

For the purpose of comparison these results may be sum- 
marized as follows: 


Average area of dermal surface innervated by one afferent fiber. 


Body as a whole, 2.05 mm?. 
Head and Neck, 1.08 ‘<< 


Arm, eROnee: 
Leg, Pegs, 0 
Trunk, SUG es 


4. Comparison and Discussion. As already shown, 
DoNALDSON’s estimate of the innervation of the skin is, on 
the average, one cutaneous fiber to every 6.3 mm”. of the 
surface of the skin. Since this estimate is based on Stit- 
LING’s determination of the number of nerve fibers in the 
dorsal spinal roots, it is evident that the source of the differ- 
ence between this estimate on innervation and my own is 
the fact that STILLING’s results are only 40% of mine. To 
show that DoNALDsON anticipated that STILLING’s results would 
prove less than they should be, I quote his statement concern- 
ing them: ‘‘It seems probable that both these estimates (i. e., 
for the fibers in the dorsal and ventral roots) were too low” 
(DoNALDSON 1901). 

The difference between VoISCHVILLO’s results and my own 
is due chiefly to the fact that he used 130,084 mm’. and 
303,566 mm’. respectively for the area of the dermal surface 
of arm and leg, while I used for the same 199,860 mm’. and 
501,680 mm’. In other words, using equal values for the area 


INGBERT, Cutaneous Innervation in Man. 219 


of the dermal surface, the estimates for the cutaneous innerva- 
tion by VOoISCHVILLO will be very nearly the same as that ob- 
tained by myself. 


A study of the following tables makes this apparent : 


Estimate by VOISCHVILLO. 


Area of Dermal Surface 


Dermal area Number of innervated by one cuta- 
Part in sq. mm. Nerve Fibers neous nerve fiber. 
Arm 130,084 119,337 I.1 mm? 
Leg 303,566 154,459 1.9 mm? 


Estimate by Author. 


Area of Dermal Surface 


Dermal area Number of innervated by one cuta- 
Part in sq. mm. Nerve Fibers neous nerve fiber. 
Arm 199,860 152,970 1.30 mm? 
Leg 501,680 204,785 2.45 mm? 


Estimate based on Number of Fibers used by VoiscuviLLo and area 
of dermal surface used by Author. 


Area of Dermal Surface 


Dermal area Number of innervated by one cuta- 
Part in sq. mm. Nerve Fibers neous nerve fiber. 
Arm 199,860 119,337 1.42 mm? 
Leg 501,680 154,459 2.25 mm? 


One factor, however, is not considered in this comparison, 
viz: VOISCHVILLO’s observations being taken at the periphery of 
the body would tend to give him a high number of fibers 
owing to their branching. But as his estimate onthe innervation 
(using the same dermal areas) is very nearly the same as 
that obtained by myself, it is likely that his estimate of the num- 
ber of cutaneous fibers is too low. This agrees with the fact that 
the calculation of the cutaneous nerve fibers in the dorsal roots 
of the spinal nerves based on VoIscHVILLo’s data, gives only 
60 to 67% while that based on my own data gives 79.22%. 

Concerning my own estimate for the innervation of the 
skin, it may be remarked that it is probable that in different 
persons the skin areas differ more than the numbers of nerve 
fibers in the dorsal roots of the spinal nerves, and since I have 
made use of the spinal roots of a large man and the largest 
area of the dermal surface on record, it is probable that if a 


220 JOURNAL OF COMPARATIVE NEUROLOGY. 


smaller subject were used for the determination of both the 
number of fibers and area of skin, the area of skin for each 
cutaneous nerve would be somewhat smaller. 


V. THE BEARING OF THE AUTHOR’S ESTIMATE OF THE INNERVA- 
TION OF THE DERMAL SURFACE ON THE THEORY OF THE SPE- 
CIFIC ENERGIES OF NERVES. 


It might be of interest to consider the bearing this esti- 
mate of the innervation of the skin has upon different theories 
as to the number of classes of nerve fibers that mediate the 
dermal sensations. If we assume with WEBER (1846) that 
impulses giving rise to sensations of heat, cold, pressure, and 
pain, pass over the same afferent cutaneous nerve fiber, then 
the above calculations will hold true, viz., one cutaneous nerve 
fiber, on the.average, innervates 2.05 mm’. of the dermal 
surface, giving all these forms of sensation to that area. If, on 
the other hand, we assume with Foster (1891) that there are 
four classes of afferent cutaneous nerve fibers, then this esti- 
mate will have to be so changed that one cutaneous nerve fiber 
of each class will innervate, on the average, 42.05 mm7?., or 


8.2 mm’. of the dermal surface. To specify more in detail, 
one cutaneous nerve fiber of each class will have to innervate 


4X1.3 mm’*., or 5.2 mm”. of the dermal surface of the arm; and 
4X3.15 mm”, or 12.6 mm’. of that of thetrunk. If this theory 
be true, then a histological examination of the nerve termina- 
tions in the skin ought to show each cutaneous fiber of the dor- 
sal roots to innervate, on the average, an area of the skin of 
the trunk equal to 12.6 mm*. However, until we know the 
amount of branching of these fibers, not only in the skin but 
also in the peripheral trunks, we are unable to judge whether 
or not it is possible for one nerve fiber to innervate so large an 
area of the skin. 
VI. Summary. 

1. According to the estimate here made, about 79% of 

the medullated nerve fibers in the dorsal roots of the spinal 


nerves of both sides, or 1,032,730 fibers, go to innervate the 
dermal surface and about 21%, or 274,521, are afferent fibers 


INGBERT, Cutaneous Innervation in Man. 221 


distributed to muscles and deep tissues. The afferent fibers of 
spinal ganglion origin passing in the vam communicantcs are not 
separately considered in this estimate, but for the moment are 
classed with those passing to the skin. 

2. According to my estimate (using the skin areas fora 
large man), one cutaneous nerve fiber in the dorsal spinal roots 
innervates, on the average, 1.08 mm’. of the skin of the head 
and neck, 1.30 mm’. of the skin of the arm, 2.05 mm” of 
the skin of the entire body, 2.45 mm”. of the skin of the leg, 
and 3.15 mm. of the skin of the trunk; and for each addi- 
tional class of nerve fibers assumed we must increase the area 
proportionately. 

3. If we assume, with Fosrer, four classes of cutaneous 
nerve fibers, then each fiber will have to innervate, on the 
average, 4.32 mm’. of the dermal surface of the head and neck, 
and 12.6 mm’. of the dermal surface of the trunk. 

4. If there be four classes of afferent nerve fibers in the 
dorsal roots of the spinal nerves of man, then a histological 
examination of the nerve terminations in the skin ought to 
show each cutaneous nerve fiber to innervate, on the average, 
areas of skin as large as given above. 


VII. BIBLIOGRAPHY. 
Donaldson, H. H. 
1901. Am. Text-Book of Physiol.. Phzladelphia, and London, Vol. Il, 
p. 230, 
Foster, M. 
1891. Physiology, Pt. 1V, London, ps 281. 
Fubini, S. and Ronchi, J. 
1881. Untersuch. z. Naturl. d. Mensch, u. d. Thiere. XII, p, 26. 
Funke, O. F. 
1858. Untersuch. z. Naturl. d. Mensch, u. d. Thiere, IV, p. 36. 
Henle, J. 
1879. Handb. der Systemat. Anatomie. Braunschweig. Bd. \II, Abthe 
I, pp. 554 and 594. 
Ingbert, C. E. 
1903. Journ, Comp. Neur., Granville, Vol. XIII, p. 53. 


222 JOURNAL OF COMPARATIVE NEUROLOGY. 


Krause, C. F. 
1844. Wagner’s Handw. d. Physiol. Bd. II, p. 131. 
Meeh, C. 
1879. Ztschr. f. Biol. Bd. XVIII, p. 425. 
Rauber, A. 
1893. Lehrbuch der Anatomie des Menschen. 4. Aufl. 
Sherrington, C. S. 
1894-95. /. Physiol., London, XVII, p. 211. 
Stilling, B. 
1859. Neue Untersuchungen iiber den Bau des Riickenmarks, Cas- 
sel, p. 602. 
Voischvillo, |. (or Woischwillo). 
1883. Relation of Calibre of Nerves to the Skin and Muscles of Man. 
(Russian). St. Petersburg. 
Weber, E. H. 
1846. Wagner’s Handw. Bd. 3, Abth. 2, p. 500. 


ON A SAW. DETERMINING THE NUMBER (OF 
MEDULLATED NERVE FIBERS INNERVATING 
ia iniGh, SHANK AND) POOP OF | THE 
FROG—RANA VIRESCENS. 


By Henry H. Dona.pson. 


(From the Neurological Laboratory of the University of Chicago.) 


During the past three years, while following the studies of 
Dr. Dunn (1900 and 1902) on the innervation of the frog’s leg, 
it has been my endeavor to discover whether there was any law 
determining the distribution of the medullated nerve fibers to 
the segments of the leg. This law has been found and is 
expressed as follows : 

The nerve fibers entering the leg of the frog (Rana virescens) 
by the sciatic and crural nerves, are distributed to the thigh, shank 
and foot in numbers which, for each of these segments are equal 
to the sum of the efferent fibers,—taken in proportion to the weight 
of the muscles,—and of the afferent fibers,—taken in proportion to 
the area of the skin, 

The data discussed in the following pages are intended to 
furnish the evidence for the law just stated. 

To make this investigation it was necessary to know: 


(1) The relative number of medullated ventral and dor- 
sal root fibers in the nerves supplied to the frog’s leg. 

(2) The relative weight of the muscles of the thigh, 
shank and foot. 

(3) The relative areas of the skin for the thigh, shank 
and foot. 

(4) The number of medullated nerve fibers entering the 
leg and also the number distributed to each segment. 

(5) The number of medullated fibérs distributed as mus- 
cular and cutaneous nerves to each segment of the leg. 


224 JouRNAL OF CoMPARATIVE NEUROLOGY. 


With the exception of the data called for under (3), the 
facts needed were to be found in papers already published from 
this laboratory. We shall take up the points in the order just 
given. 

(7) The relative number of medullated ventral and dorsal 
root fibers in the nerves supplied to the leg. 

The nerves sending fibers to the leg are the VII, VIII and 
IX spinal nerves as usually numbered, or, according to the 
recent numbering of Gaupp (1897), the VIII, IX and X. 

According to the enumeration of the medullated nerve 
fibers in the dorsal and ventral roots of these nerves by Har-. 
DESTY (1899, p. 84), the proportion is. 


TABEEGE: 
Relative number of fibers in roots, 
Ventral. Dorsal. 
Frog weighing 48 grams 100 177 
ie te 59 grams 100 175 


The average of these two observations gives therefore 100 
ventral root fibers to 176 dorsal root fibers, and this is the ratio 
here employed. Such being the relation found in the spinal 
roots, it is assumed to be the same in the sciatic and crural 
nerves at the point where they enter the leg. Whatever the 
number entering of fibers the leg, they are then to be divided, 
as motor and sensory, in the above proportion. 

In this calculation no correction for possible efferent fibers 
in the dorsal roots is attempted, for we have no data with 
which to work. When efferent fibers appear in the dorsal 
roots of the frog, there is indirect evidence that the number 
must be quite small, and they are here neglected (HortTon- 
SMITH 1897, Wana 1898, Date 1901). The ventral root is 
assumed, therefore, to contain only motor or efferent fibers 
and the dorsal root only afferent or sensory fibers. 

Having determined the proportional numbers of the 
afferent and efferent fibers, the next step is to present the meas- 
urements according to which these fibers are to be distributed, 
namely, the relative weights of the muscles and the relative areas 
of the skin in the several segments of the leg. 


Donatpson, Law of Innervation. 225 


(2) The relative weight of the muscles of the thigh, shank 
and foot. 

In an earlier study, made in collaboration with Mr. ScHoeE- 
MAKER (1900) it was determined in Rana virescens that the 
relative development of the muscles of the frog’s leg, as indi- 
cated by the weight of the muscles of the thigh compared with 
those of the remainder of the leg, was nearly constant for all 
groups above five grams in body-weight (see DonaLpson and 
SCHOEMAKER, 1900, pp. 124-125). 


TABLE ILI. 


Giving in grams the weights of the muscles in the thigh, shank and 
foot of the Frog—R. virescens. (From Table VII, DonaLpson 
and SCHOEMAKER, 1900.) 


Boay Weight of Muscles 


No| Sex | Weight | Leth. | Thigh| Shank} Foot 


rigeil| oly 29.40 E85) | 3-LO5| Koray) 2533 
3-097 1.201] .538 


O| 1.240] .659 


19] -F | 30-45 179 | 3-3 
era -195| .603 


GQ Go 
ho 
| 


_ 


20| F 33-96 172 3-480] 1.393} .603 
3-681) 1.354] .631 


22| M | 38.16 200 | 3-637} 1.563) .715 
3-536] 1.456} .653 


23) ik 42.54 215 4-455| 1.562) .822 
4-369) 1.535] .843 


_ 


25} M | 45.37 205 | 4-383) 1.496] .869 
4.278] 1.493 816 


-797| 1.844] .926 
.830| 1.828} .926 


4 
4 
AA Ae 47-58 206 4. 
4 


713) Deepal. 824 

-616) 1.813} .871 

23) F 48.33 220 | 5-402; 1.906) 1 006 
§-430| 1.949] 1.019 


> 2c 52.55 206 | 5-404) 2.450)  .835 
5 456) 2.400] .850 


226 JoURNAL OF COMPARATIVE NEUROLOGY. 


From Table VII, in that paper is taken the following series 
of records comprising numbers 18-29, inclusive, (excluding 
numbers 21 and 24 for which the muscle weights were not 
determined). This is the series just as it stands in the original 
table, and these same records have also been employed in my 
paper ‘‘Ona Formula for Determining the Weight of the Cen- 
tral Nervous System of the Frog from the Weight and Length 


of its Entire Body,”’ 1902. 
For a description of the method by which the weights of 


the muscles were determined, the reader is referred to the 
original paper (p. 118). 

For the first five frogs entered in the table, calculation 
gives the following proportional values for the weights of the 
muscles of the thigh, shank and foot: 


TABLE III. 


Segment of Leg 


Percentage Value of Weight of Muscles. 


Thigh -63-9% 
Shank 24.3% 
Foot 11.8% 


In the case of the second five frogs in this same table, the 
proportional values are as follows: 


TABLE, LY: 
Segment of Leg, Percentage Value of Weight of Muscles. 
Thigh 63.9% 
Shank 245% 
Foot 11.6% 


It will be seen that these two series differ only by some 
tenths of a percent. in the case of the shank and foot, so that 
the relative values may be considered fairly constant. 

For use in the present investigation we take the average of 


the two series which gives: 


TABLE V. 
Segment of Leg. Percentage Value of Weight of Muscles. 
Thigh 63.9% 
Shank 24.4% 
Foot 11.7% 


Having obtained the data according to which the motor 


fibers entering the leg should be distributed, we need next to 


Donatpson, Law of [nunervation. 227 


obtain the corresponding data for the areas of skin, —in accord- 
ance with which the afferent fibers are to be distributed. 

(3) On the relative areas of the skin of the thigh, shank, 
and foot. 

Since the relative weights of the muscles in the several 
segments of the frog’s leg remain unaffected by the size of the 
frog (DonaLpson, 1898, and DonaLpson and SCHOEMAKER, 
1900), and since the conformation of the leg is similar in large 
and small frogs, it follows that the relative areas of the skin 
covering the different segments of the leg are unaffected by the 
absolute size of the frog examined. This removes the neces- 
sity of always working with frogs of the same size. 

My acknowledgments are due to Dr. Dunn for working 
out the area of the skin in the several segments of the frog’s 
leg. As the method used was somewhat novel, it will be nec- 
essary to describe it in detail. 

The frogs examined had the following body measurements: 


TABLE VI. 


Frogs examined for area of skin. 
Length from tip of nose to 


Frog No. Body Weight, grams. end of longes !toe. 
g y ght, ¢ g 

I 34.76 190 mm. 

Zz 36.75 195 mm. 

B 28.17 187 mm. 


The procedure was as follows: The frog was killed with 
chloroform and pinned out on a sheet of cork with the ventral 
aspect uppermost. One leg was extended and abducted so 
that it made an angle of about forty-five degrees with the longi- 
tudinal axis of the body. A strip of glass about 3 cm. wide 
and a few centimeters longer than the frog’s leg, was then 
passed beneath it. The leg thus rested on this strip of glass 
and was in turn covered by a similar strip of the same size. 
On the upper surface of the latter a layer of tracing paper had 
been fastened by a drop of paste at each of the four corners. 
Rubber bands were next passed over the two strips of glass 
and the leg lightly compressed between them. Before putting 
on the covering strip, however, the foot was stretched so as to 


228 JOURNAL OF COMPARATIVE NEUROLOGY. 


exhibit the full extent of the web. For this purpose, threads 
were tied to the first, third and fifth toes, pulled taut and fast- 
ened by pins. The web was stretched until the second toe 
showed a tendency to curl up, and this reaction was regarded 
as indicating the normal extension of the web. When so 
extended, the general outline of the foot is that of a quad- 
rilateral. 

These adjustments having been made, it was determined 
by measurements taken between the two strips of glass, that 
their faces were approximately parallel. Next an outline, 
just outside the contact line of the upper glass with the skin of 
the leg, was made with a pencil on the tracing paper, starting 
from the uppermost part of the thigh on one side, and follow- 
ing around the entire outline of the leg to the uppermost part 
of the thigh on the opposite side. The upper limit between 
the skin of the body and that of the thigh was marked by the 
adjustment of the glass strips, and the levels of the knee and 
ankle joints respectively, were fixed by passing needles through 
these joints when preparing the leg for this examination. By 
drawing, therefore, a line across the tracing paper at these 
several levels, the outlines for the thigh, shank and foot were 
obtained. It was assumed that the outline on the dorsal aspect 
of the leg would be similar to the one found on the ventral 
aspect. To obtain the areas for the sides of the leg, the dis- 
tance between the two glasses at the hip, knee and ankle was 
measured and the length of the curved line (the ‘‘coast line’) 
forming the lateral boundary in the case of the thigh and 
shank, taken on each side, by means of a waxed thread applied 
to the outline. This gave the initial measurements for the cal- 
culation of the area of the skin of the thigh and shank. Con- 
cerning the foot we shall speak later. 

To illustrate the determination of the area for a given 
segment, let us take the thigh. In this case we have the area 
marked on the tracing paper between the boundary separating 
the thigh from the body and that separating it from the shank. 
The content of this area was determined in square millimeters 
by means of a planimeter and then doubled so as to include the 


Donatpson, Law of lnnervation. 229 


corresponding area on the opposite side; to this sum was 
added the areas obtained by multiplying the length of each 
curved side (the ‘‘coast-line’”’), by a number equal to half the 
sum of the height of the side at the hip plus its height at the 
knee. 

The sum of the four areas thus obtained was considered as 
the total area of the skin of the thigh. By following the same 
steps, the total area of the skin of the shank was found. On 
coming to the foot, however, the method of determining the 
area required to be modified in order to include the additional 
surface represented by the toes. The original outline on the 
tracing paper gave the area of one side of the foot projected 
on a plane surface, to which, when doubled, should be added 
the amount represented by the elevation of the toes and that 
due to the thickness of the tarsus. 

To illustrate the method used, the following figure is 
given: | 


230 JoURNAL OF COMPARATIVE NEUROLOGY. 


Knee 


Ankle 


Figure 7. Outline of Frog’s leg, showing the outline of the ventral aspect 
of right leg. AHi#p—marks the junction line of the body with the thigh. 
Knee—the junction line of the thigh with the shank. Am&/e—the junction 
line of the shank with the foot. D’—D.—length of tarsus plus second toe. 
M-A—length of fifth toe. A/’—#.—length of fourth toe. M’’—C—length of 
third toe. (Second toe measured with tarsus). J/ ’’—#—length of first toe. 


Donatpson, Law of /nunervation. 231 


The length of the line D/-D was first taken, then the 
thickness of the foot at D’ and at D. Half the sum of these 
last two measurements multiplied into the length D’—D gave the 
lateral areas of the foot, considering these as extending from 
the ankle joint to the tip of the longest toe, and twice this 
gave the areas for both sides. The length of the lines JA, 
M'B, M'' Cand M''' Ewere then taken, and the thickness 
githe, footcat Wi, 07’, A7'', M'"! and at ‘each ‘of the” toes 
A, B, Cand £ (D being omitted, as it had been already meas- 
ured). Half the sum of these two thicknesses was taken in 
each case as the average thickness of the toe, and the product 
of these numbers into the length of the respective toes, gave 
the area of the skin on one side of the toe considered as a 
plane surface. The surface is so rounded, however, that about 
one-half of the area thus calculated has already appeared in the 
surface projection of the foot, and has been counted in the 
area incluced within the original outline; therefore, intstead of 
taking this area for the side of the toes twice, it was taken 
only once. 

Thus to get the total area for the foot, there are added 
together: the two areas from the tracing—the number of 
square millimeters being determined by the planimeter—the 
area for the longest toe including that for the sides of the tar- 
sus and finally the additional areas for the four remaining toes. 

The areas thus obtained for the three frogs measured are 
given in the accompanying table, the records for the two legs 


being entered separately. 

TABLE VII. 

Giving the areas in square millimeters of the skin covering the thighs, 
shanks and feet of three frogs—R. virescens. The method of 
measurement is given in the text. 

Areas in Square Millimeters. 


— 


Frog. Thigh. Shank. Foot. 
1 Right 1584.1 1136.8 1695-3 
1 Left 1675-5 1059.0 1793 6 
2 Right 1519.2 1071.5 1539-1 
2 Left 1506.3 1035-0 1538.7 
3 Right 3177/0) 1010.9 1396.5 
2) Left 1356.2 972.4 1434.1 


Totals 9018.3 6285.6 9397-3 


232 JOURNAL OF COMPARATIVE NEUROLOGY. 


‘When the percentage values of these numbers are found, 
they give the following: 


TABLE VIII 
Average for first Average for second Average for all. 
Segment. 3 entries 3 entries 6 entries. 
Thigh 36.5% 36.5% 39.5% 
Shank 25.0% 25.8%, 25.4% 
Foot 38.5% 37-7% 38.1% 


On attempting to apply these percentages in the calcula- 
tion for the supply to the leg, it was found that the numbers 
obtained were somewhat larger than we expected them to be 
in the case of the thigh. It appeared that the area for the 
thigh had been over-estimated. Dr. Dunn therefore undertook 
a reinvestigation of the innervation of the skin of the thigh. 
This showed that the area was over-estimated in this sense, that 
in addition to the fibers entering the leg by the sciatic and 
crural nerves. there was a small independent bundle of nerves 
which entered through the anus and was distributed to a small 
triangle of skin just behind the anus and on the mesal surface 
of the leg. The area of this piece of skin thus innervated, 
was found to be 2.5% of the area of the entire thigh. For 
the details of this determination the reader is referred to a 
forthcoming paper by Dr. Dunn. 

To control these results, other measurements were made 
by an entirely different method. A plaster mould of the frog’s 
leg was taken and the leg then cast in Woops metal. On this 
cast the thigh and shank were examined for their respective 
areas by winding them carefully with fine copper wire—of a 
uniform diameter. The diameter being constant, the areas 
covered by the wire would be directly proportional to the 
length or weight of the wire used. The relations were tested 
by weighing the wire. The weight of the wire needed to 
envelop the thigh, was to that needed to envelop the shank 
as 1.43: 1. 

On comparing this ratio with that from the areas for the 
thigh and shank, as shown in Table 7, it appears that this ratio 
is also 1.43: I. 

Thus the proportional areas for the thigh as determined 


Donatpson, Law of Innervation. 233 


by these two methods are identical. It would follow from 
this that the first method was probably accurate. On account 
of its shape the foot cannot be tested by this latter method. 
When the area of skin for the thigh is reduced by the 
amount which is not innervated by fibers entering the leg in 
the sciatic and crural nerves, the following percentages are 
obtained as a general average from the three frogs measured : 


TABLE IX. 
Area of Skin. 
Thigh 35.9% 
Shank 25.7% 
Foot 38.4% 

The above percentages are those used in the calculations 
which follow. 

(4). The number of medullated nerve fibers entering the leg 
and also the number distributed to cach segment. 

Having thus determined the proportion in which the ven- 
tral root fibers and the dorsal root fibers entering the leg should 
be distributed to the several segments, it becomes desirable to 
estimate by means of the preceding tables, the total number of 
fibers going to each segment. 

To show how this is done, let us assume that 100 ventral 
root fibers and 176 dorsal root fibers enter the leg—these num- 
bers are in the proportion which has been determined. Then 
63.9% of the 100 ventral root fibers go to the thigh and 35.9% 
of the 176 dorsal root fibers also go to the thigh. Now, in 
order to determine what percentage of the total 276, the sum 
of these two numbers is, we should divide their sum, namely, 
63.9 fibers+63.2 fibers = 127 fibers by the total number of 
entering fibers, namely, 276. We then find that 46% of all 
the entering fibers go to the thigh when they are distributed in 
the above manner. 

Extending the calculation in the same way to the shank 
and foot, we may tabulate the results as follows: 


234 JoURNAL OF COMPARATIVE NEUROLOGY. 
TABLE X. 


Hereafter designated as the ‘‘formula’”’. 


46% go to the thigh 
25.3% go to the shank | eer shank 46.8% 
28.7% go tothelfoot ) {my Jo foot 53.2% 


100.0% go to the leg. 


If 46% go to the thigh, then 54% go to the remainder of 
the leg, i.e., shank and foot combined. Further, if we consider 
alone the fibers which go to the shank and foot and express 
the number as 100%, then of this 100% going to the shank 
and foot, 46.8% go to the shank and 53.2% to the foot. 
These calculations of the fibers distributed to the shank and 
foot are introduced here as they will be used later on. 

The values inthe above Table X are those with which 
all the subsequent calculations are made. It remains now to 
present the data concerning the number of nerve fibers 
observed to enter the leg of the frog and then to see in how far 
the estimated numbers for the segments of the leg correspond 
with the numbers which have been observed. 

These observed numbers have been obtained by Dr. 
Dunn. The designation of the frog and the segment or seg- 
ments for which the enumerations have been made, together 
with the date of publication, are given in the following Table 
XI in chronological order: 


TABLE Xl. 
Frog B Thigh DUNN 1900 
Frog C Thigh DUNN 1900 
Frog B II Thigh, Shank and Foot DUNN 1902 
Frog C II Thigh, Shank and Foot DUNN unpublished 


Below are given the details of the data for each frog so far 
as they will be required in this study. 

The accompanying Figure 2, taken from Dr. Dunn’s 
paper (1902), shows the levels at which the nerves for the 
frog’s leg were sectioned and where the number of fibers was. 
counted. 


Donatpson, Law of Innervation. 235 


> Ol D. Cre 


al 


Sat aa dal 


) 


Figure 2. Diagram of the innervaticn of the Frog’s Leg, showing the 
levels at which the number of nerve fibers was counted. C—level of Crural 
nerve just above branches to thigh. Sv—level of Sciatic nerve just above 
branches to thigh. Sz2—=level of Sciatic nerve just below branches to thigh. 
P=level of Peroneal nerve just above branches to shank. Z=level of Tibial 
nerve just above branches to shank. Pr, P2=level of Peroneal nerve just 
above branches to foot. 77, 72=Ilevel of Tibial nerve just above branches 
to foot. 


236 JOURNAL OF COMPARATIVE NEUROLOGY. 


Thus all the fibers entering the leg were in the sections at 
the levels C+S,. All the fibers entering the shank were in the 
sections at P+T, and all the fibers entering the foot at P1+P2 
and T1+T2. 

In the Tables XII-XVI which follow, these designations 
are used to indicate the levels at which all the fibers for a given 
segment were to be found. Below the entry for the number of 
fibers to each of the segments, stands ‘‘To Thigh” and ‘‘To 
Shank,’ which means that all the branches to the thigh or to 
the shank contained the numbers there entered, under 
‘‘observed.”’ The number of fibers immediately below the 
branches to the thigh is also given after Su». 

In the column headed ‘‘ muscular and cutaneous,”’ are the 
observed numbers found in the muscular and cutaneous 
branches and their sum is equal to the ‘‘ observed number ’”’ in 
each case. 

In the column headed ‘‘calculated,’’ we have the number 
determined by Dr. Dunn by taking the difference between the 
number just above the branches to the thigh (C+S, ), or shank 
(P+T), and the number just below at S: or at Pi+P2 and 
ash: 

In the column marked ‘‘splitting fibers,” we have the 


’ 


difference between the observed—the larger number—and the 
calculated—the smaller number—which difference is due to 
fibers that have split in their course so that they are repre- 
sented either by two divisions having the same distribution 
(i. e., both run in the branches supplying the segment), or by 
one division running in the branches to the segment, and by 
the other in the main trunk of the nerve to some point below 
the Jevel of the branches. It is the presence of these splitting 
fibers which renders complicated the application of the test 
which it is proposed to make, since the formula based on the 
law of distribution does not take account of the additional fibers 
which appear as the result of splitting. 


DonaLpson, Law of Innervation. 


237 
TABLE XII. 


Data taken from Dr. DuNn’s paper, 1900; Tables III, IV, V and VI. 


Frog B. 


Locality 


C+S, 
To thigh 


Sy 


C+5, 
To thigh 


S, 


Right Leg. 


Left Leg. 


Number of Fibers 
Observed Calculated Splitting Fibers 
Cutaneous = C 

sare) Muscular = M 

0 
C ( 876 

2623 2405 218 
M (1747 

28680 eases yi Sees e Sy ee 

5385 0-H ----  we ne 
CO (RS 71 

2600 2411 189 
M (1729 

2074. weneee een wee 


TABEE XIII. 


Data taken from Dr. DuNnn’s paper, 1900; Tables III, IV, V and VI. 


Frog C. 


Locality 


c+5, 
To thigh 


S, 


C+5S, 
To thigh 


S, 


Right Leg. 


Left Leg. 


Number of Fibers 


Observed Calculated Splitting Fibers 


Cutaneous = C 


Total Muscular = M 


6011 


--<-<=<= @-—-<-<=<= =eeen= 


238 JOURNAL OF COMPARATIVE NEUROLOGY. 


TABLE XIV. 


Data taken from Dr. Dunn’s paper, 1902; Tables V to XIV inclusive. 


Frog B II. 


Locality 


C+5S, 
To thigh 


S, 
| ao 


To Shank 


Pi -+ P2and T1-+ T2 


Right Leg. 


Number of Fibers 


Observed 


Cutaneous = C 
Total Pee = M 


rite iak pee eaee 


TABLE XV. 


Calculated 


Splitting Fibers 


Data taken from Dr. DuNnn’s paper, 1902; Tables V to XIV inclusive. 


Frog B II. Left Leg. 


Locality 


c+5S, 
To thigh 


Sa 
PT 


To shank 


Pr-+ P2and Ti + T2 


Number of Fibers 


Observed 


Cutaneous = C 


Total Muscular = M 


ie ak oo 


Calculated 


Splitting Fibers 


DonaLpson, Law of lnnervation. 239 
TABLE XVI. 


Data from specimen of R. virescens. Body-weight 51 grms. Ob- 
servations by Dr. DUNN not previously published. The section 
at S; of the left leg was unfortunately too imperfect to be counted. 


Frog C II. . Right Leg 
Number of Fibers 


Locality Observed Calculated Splitting Fibers 


f Cutaneous = C 
oe \ Muscular = M 


1301 C 
an lees Si 

OMIM Aye eos ty 0M) Oe ee te MUNG hy reece 
Sy S338 a ———— ee 


a eee eee 2a OZ yw eee! Bo ee eee Sh tt Ue ee 


Left Leg 
- = 1323,€ 
See { vals S Si (section lost) soa2n— es einem 
S, Bg800 ieeecoeaeyne a ienoamoe 1) AM RE one 
IPR te Jes ahavel, Wi Aad, ikolsso) (= SER eke eet 


To illustrate the use of the data in the foregoing tables, 
we will take the right leg of Frog B II (Table XIV), and 
determine how far the observed numbers there given agree with 
those estimated by the aid of my own formula. Table XIV 
shows 7107 fibers entering the leg. When the proportional 
numbers are calculated according to the formula (Table X), we 
obtain the following: 


TABLE XVII. 


Entering right leg of Frog B II— 7107 fibers 
To thigh— 46.0%= 3269 fibers 
To shank— 25-3%= 1796 fibers 
To foot— 28.7%= 2042 fibers 


The above numbers of fibers are the numbers of single 
pathways between cach segment of the leg and the entering nerve. 

When the observed numbers of fibers to the several seg- 
ments of the leg (see Table XIV) are compared with those 
here estimated, we obtain the following: 


240 JOURNAL OF CoMPARATIVE NEUROLOGY. 


TABLE XVIII. 
Number of fibers to the several segments of the right leg of Frog B IT. 


Estimated Observed 
(DONALDSON) (DuNnN) 
To thigh 3269 3508 
To shank 1796 2130 
To foot 2042 2497 


Thus in every case the numbers estimated by the formula 
are less than those observed by Dr. Dunn. But as Table XIV 
shows, Dr. Dunn’s count includes a large number of fibers 
which have split (see ‘‘splitting fibers” in tables), and which, 
therefore, are not taken into account by the formula, which 
applies to single pathways only. 

The next step, therefore, in the comparison is to bring the 
two series of numbers to the same standard. There are two 
ways of doing this, namely, either by removing the additional 
fibers due to splitting from the number observed by Dr. Dunn, 
or by adding the additional fibers to the number as determined 
by the formula. The latter method was followed; first, 
because it was desirable to consider the odserved number as the 
fixed standard, not to be altered in any way, and second, 
because it seems probable that the most direct utilization of 
this formula will be for determining the number of fibers which 
a direct count would show to be supplied to a given segment. 

Thus in the further work, the observed number is in each 
case taken as the standard, and the calculations are directed to 
determining the number which must be added to that estimated 
by the formula in order to make it comparable with the stand- 
ard. To illustrate the stéps taken for this purpose, we shall do 
best to examine the innervation of the thigh in the case chosen, 
namely, the right leg of Frog B H. My estimation gives for 
the thigh 3269 fibers and Dr. Dunn’s observation 3508. 
When, however, in this instance the number of fibers observed 
above the branches to the thigh at (C+S:) was diminished by 
the number observed just de/ow the branches at (S:), the 
difference was found by Dr. Dunn to be 343 (see Table XIV). 
Thus her observed number exceeded her number calculated in 
this manner by 343 fibers. This excess is explained as the 


Donatpson, Law of Innervation. 241 


result of splitting fibers and means—if dichotomous splitting 
be the form assumed—that 343 of the fibers counted at C+S, 
have split later into two divisions each. Just where this split- 
ting occurs, we are obliged to guess. The numerical results 
indicating the number of splitting fibers would be the same, 
whether it occurred in the main stem, C+Si, just where the 
branches to the thigh are given off, or within the branches after 
these had been given off, but before the levels at which they 
were individually sectioned. For the further development of 
our idea of the innervation of the leg, however, these two pos- 
sible arrangements leading to like numerical results, have very 
different values. In the latter instance, where the splitting 
occurs within the branches, the splitting fiber is in no wise rep- 
resented in the main trunk of the nerve below the level of the 
branches, whereas in the former instance, where splitting occurs 
in the main trunk, one division can appear in the branches and 
the other. continue in the trunk, the original fiber being there- 
fore still represented in the trunk, below the point at which it 
has split. 

It was necessary to assume that some of the fibers that 
split did so in the trunk, and thus were represented both in the 
branches and in the trunk below the branches, while others 
split in the branches themselves; both divisions in this case 
having a similar distribution. 

After several tests, it was found that the most accordant 
results were obtained when it was assumed that one-third of the 
splitting fibers split in the branches. Fibers splitting in this 
way form Class a. Since the total number of splitting fibers 
was in this case 343, one-third would be 114.3 fibers. These 
are to be added to the number estimated by me (see Table 
EGVAI 3200 C114. 3-3 333. 3: 

The number of fibers that pass to the shank and foot is 
not affected by this process, since each one of these fibers split- 
ting in the branches is without representation in the trunk 
below the level of the branches. 

By this step 114.3 of the splitting fibers have been with- 


242 JOURNAL OF COMPARATIVE NEUROLOGY. 


drawn from the original 343, leaving 228.7 to be still 
considered. 

These remaining fibers are assumed to be of the second 
class of splitting fibers (Class b), i. e., those, one division of 
which passes into the branches, while the other continues in 
the trunk below the branches. 

Since by the formula we have taken 46% of the total 
number of the fibers entering the leg (see Table XVII), we 
must assume that we took 46% of the fibers with double rep- 
resentation (Class b). It follows, therefore, that in the original 
46% or 3269 fibers, there are already included 46% of the 
fibers with double representation, or 46% of the 228.7 fibers 
which constitute Class b at this level. Therefore, to bring the 
estimated up to the observed number in this respect, there are 
still to be added to the estimated number 100%—46%=54 % 
of 228.7, which equals 123.5 fibers. In general, therefore, to 
make the estimated number comparable with that observed by 
Dr. Dunn, we should add one number representirg the fibers 
splitting in the branches (Class a), and also a second number 
representing the splitting fibers with double representation 
(Class b), in so far as they have not been included in the 
original 46%. 


TABLE XIX. 


Showing the number of fibers to be added to the original 46% in the 


case of Frog B II—Thigh. 
Calculated numbers. Observed number 


Original 46% == 209 
Fibers splitting in branches (Class a) = AGS 
Splitting fibers with double repre- 
sentation (Class b) eRe s 
3506.8 
or 3507 3508 
Difference = I fiber 


As the table shows, the estimate is almost identical with 
the observation, being but one fiber less. Without comment- 
ing on this result at this place, we shall pass on to the consider- 
ation of the nerve supply to the shank of this same leg. 

It must be kept clearly in mind that our calculation for the 


Donatpson, Law of Lnnervation. 243 


percentage distribution of the fibers to the segments of the leg 
applies to the number of szxzgle medullated fibers found entering 
the leg at C+S; in this case, 7107 fibers. Having removed 
46% of this number for the innervation of the thigh, there 
remain 54% to be distributed to the shank and foot. 

Thus: 7107—3269=3838 fibers. 

The law calls for a distribution of this 54% of the initial 
number in the proportion of 25.3% to the shank and 28.7% 
to the foot. When, however, we attempt to deal with the 
splitting fibers belonging to the shank and foot in making the 
estimated comparable with the observed results, it is necessary 
to distribute these splitting fibers in the same proportions in 
the branches to the shank as they were distributed in the 
branches to the thigh, and in order to make the distribution in 
a like manner, it is necessary to treat the number of fibers 
going to the shank and foot, 3838, as 100% of all the fibers 
concerned, and then designate those to the shank as 46.8% of 
the same number, 

Since 46.8% of 3838=25.3% of 7107=1796 
53-2% of 3838=28.7% of 7107=2042 

According to the formula, therefore, 1796 fibers (single 
pathways) are estimated as going to the shank. Dr. DuNnN 
observed 2130 fibers to the shank, and further, by the same 
methods as were employed in the case of the thigh, she deter- 
mined that there were 465 splitting fibers in the branches to 
the shank. (See Table XIV.) 

In adjusting the estimated number to that observed, these 
splitting fibers are treated in the same manner as those found 
in the supply to the thigh, i. e., one-third are taken as 
dividing in the branches, and of the remainder it is assumed 
that 46.8% are included in the original estimate, so that 53.2% 
only need to be added to bring the estimate to the same stand- 
ard as that of the observed number. Thus, imitating Table 
XIX, for the thigh, we have the following: 


244 JOURNAL OF COMPARATIVE NEUROLOGY. 


TABLE XX. 


Showing the number of fibers to be added to the original 25.3% in 
the case of Frog B II—Shank. 


Calculated numbers. Observed number 


Original 25.3% of 7107 or 46.8% of 3838 = Tos 
Fibers splitting in branches (Class a) = 155 
Splitting fibers with double representa- 
tion (Class b) = 165 
2116 2130 
Difference —= —1Z¢4 fibers or 0.6% 


The approximation between the estimated and observed 
numbers is not so close here, but is still fair, being within less 
than one per cent. 

The number of fibers to the foot has still to be considered. 
In the case of the foot, the observed numbers are taken from 
the enumeration at the levels P1+P2 and T1+T2. Figure 2. 
These levels would correspond to those just above the branches 
in the case of the thigh and shank. We have to assume, 
therefore, that in the number observed by Dr. DuNN at these 
levels as going to the foot, there are in addition to the single 
pathways represented by fibers that pass unsplit from their 
entrance into the leg, two other classes, those namely (a) that 
have split within the trunk, both divisions descending to the 
foot, and (b) all other split fibers, one division of which goes 
to either the thigh or shank, while the other goes to the foot. 

Of the class (b), there are in the entire leg at least three 
groups: bi, the fibers which split, one division passing to the 
thigh and the other to the foot; bs, fibers which have split, 
one division going to the thigh and the other to the shank; 
bs, splitting fibers which have one division going to the shank 
and one to the foot. 

These several groups are all marked, Figure 3, as in bi, by, 
bs. At the level where the nerves enter the foot, it is neces- 
sary to make allowance for groups bi and bs of class (b), as 
well as for representatives of class (a). 

According to formula, 2042 fibers pass into the foot. Dr. 
Dunn observed 2497. To make the estimated number com- 
parable with that observed, the allowance for the splitting fibers 


Donatpson, Law of Innervation. 245 


must be introduced. The principle involved in making the 
allowance is that the division of any splitting fiber which has 
already been counted asa fiber entering the leg is to be con- 
sidered as additional, and hence added to my estimate of the 
number to the foot. 

For example, the splitting fibers included in the original 
46% of those entering the leg and sending one division to the 
thigh (bi) are pictured as sending the other division to 
the foot. 

This group (b,) contributes 106 fibers to the foot. As 
the fibers representing these divisions have already been ac- 
counted for, they would not appear in my estimate, and hence 
must be added to it. In like manner, the long divisions of the 
splitting fibers which go to the shank come from fibers already 
counted and are therefore additional; thus these also should be 
added to my estimate. The latter group (b,) represents 
46.8% of the splitting fibers which pass to the shank and these 
amount to 145. To the original estimate of 2042, therefore, 
106+145=254 is to be added, making a total of 2293 fibers. 
This number is still 204 fibers less than the observed 2497. 
The difference may justly be credited to the splitting fibers of 
class (a), namely those, both divisions of which have a like 
distribution. That many such splitting fibers occur is shown 
by Dr. Dunn’s observations (1902, p. 314). 

The arrangement of the fibers as here imagined is exhib- 
ited in Figure 3, which is purely a schema. Here the largest 
column represents the fibers going to the thigh, the next 
largest those to the foot and the smallest those to the shank. 
The size of the columns is proportional.to the number of fibers. 
they contain. 


246 JOURNAL OF CoMPARATIVE NEUROLOGY. 


FIBRES 
ENTERING LEG 
100% 


THIGH FooT SHANK 


oa 


8.7% 


To Foot 


Figure 3. The explanation of this schema intended to illustrate the arrange- 
ment of the splitting fibers distributed to the leg of the frog, is given on the 
following page. 


Donatpson, Law of Innervation. 247 


All the fibers entering the leg by the crural and sciatic 
nerves are enclosed within the circle marked ‘‘Fibers entering 
the leg, 100%.’’ Atthe next lower level, the small circles 
enclose all the fibers to each of the three segments of the leg 
marked respectively : 

Thigh 46.0% 
Shank 25.3% 
Foot 28.7% 

Within the last three columns the heavy black lines indi- 
cate the splitting fibers, which according to the description in 
the text, are distributed to the the thigh, shank and foot. These 
splitting fibers are divided into two classes : 

Class a—splitting fibers, both divisions of which are dis- 

tributed to the same segment : 

Class b—splitting fibers, the divisions of which are dis- 

tributed to different segments. 

This latter class is represented by three groups: 

bi, one division of which goes to the thigh and the other 
to the foot ; 

b,, one division of which goes to the thigh and the other 
to the shank; 

bs, one division of which goes to the shank and the other 
to the foot. 

C + Si, indicates the level above the branches to the thigh; 
P + T, the level above the branches to the shank, and P1 + P2 
and T1 + T2, the level above the branches to the foot. 

Within each column is represented the several sorts of 
splitting fibers, each one bearing the designation used to indi- 
cate it in the text. The designation ‘‘a’”’ always indicates the 
splitting fibers, both divisions of which have the same distribu- 
tion. Class ‘‘b”’ is represented by the three groups, b,, one 
division of which goes to the thigh, while the other goes to 
the foot ; bs, one division of which goes to the thigh and the 
other to the shank, and finally, b;, one division of which goes 
to the shank and the other to the foot. The splitting fibers are 
counted only once, and are credited to the column in which 
they appear above C+S, in the figure. The other divisions are 
in addition to the number of fibers called for by the formula. 


248 JOURNAL OF COMPARATIVE NEUROLOGY. 


The Jevels C+-Si,’P+T; and Pr+P2 and i112 are alam 
cated. If there were no splitting fibers the schema would be 
made up of the outlines of the three columns alone. 

Turning now to general results of this test, we find that in 
the case of the right leg of Frog B II, the estimated numbers 
agree very closely with those observed when an allowance, 
according to a fixed procedure, is made for the splitting fibers. 

The significance of this result depends largely on the 
possibility of repeating it. Using exactly the same methods, 
the results for the left leg of this frog (data presented in Table 
XV) are almost as good as those for the right. 


The following table presents the results for both legs 
Frog, Bull: 


TABLE XxXI. 


Estimation 
(DONALDSON) 


Corrected for splitting Observed Difference Percent- 
fibers (DuNN) in fibers age 

Thigh R 3507 3508 — I —.03% 
Thigh L 3497 3481 +16 +-40% 
Shank R 2116 2130 —14 —.65% 
Shank L 2110 2108 +2 +.09% 
Foot R 2497! 2497 Sere = 
Foot L 2486! 2486 set’ 


1 8¢, of the observed numbers being added to represent the splitting fibers 
with a similar distribution of both divisions—see text. 


This is the only frog we have in which the enumeration at 
different levels has been complete enough to permit of this full 
test of the formula. In Frog C II, however, the endeavor 
was made to determine whether the proportional distribution of 
the fibers at different levels in the leg would correspond with 
those found in Frog B II, for it seemed very probable that if 
the two frogs corresponded in this respect, they would also 
correspond in the number of fibers supplied to each segment 
of the leg. 

The data for Frog C II give the number of fibers entering 
the leg at C4 Si the number at S», and the number entering 
the foot, at Pr P2 4:/and Tria. T2; see Table Vi 


Donapson, Law of Innervation. 249 


On adding these three numbers together and then taking 
the proportional value of each of them, we find the following : 


TABLE XXII. 


Level Right Leg Left Leg 
C+5, et 


S, 29.4% 
LS il es aa) Oa A a ee 
This calculation could not be repeated for the left leg, as 
the sections for Si were imperfect. The remaining sections were, 
however, counted at levels C, S: and P1 + P2+T1+ T2 on the 
left leg, and gave numbers so closely similar to those from the 
right leg, that it is highly probable that the innervation of the 
two legs agreed closely. 
The tabulation of these same numbers in Frog B II is 
given below: 
TABLE XXIII. 


Right Leg Left Leg 
c+S, 51.6% 51.8% 
5, 30.2% 30.1% 
Pi -+2Pi+T1-+ T2 18.2%, 18.1% 


When we compare these percentages with the correspond- 
ing determinations for Frog B II, in which the two legs are 
quite similar, we find that CII has .8% more in the thigh, 
.8% less in the shank and that they exactly coincide as regards 
the foot. 

If we may judge from this, then other frogs are similar to 
B II in the general arrangement of the nerve supply to their 
legs, and hence the formula would be applicable to them as a 
class. 

In support of this view, we have the results of applying 
the formula to Frogs B and C. In these cases to be sure, both 
legs were examined, but the supply to the thigh alone was ob- 
served and therefore the comparison is limited to that segment 
of the limb. 

As the data for these calculations is to be found in Tables 
XII and XIII, it will be necessary to give only the comparison 
of the estimated number of fibers in each case with those 
observed. 


250 JOURNAL OF COMPARATIVE NEUROLOGY. 


TABLE XXIV. 


Estimation (DONALDSON) 


Corrected for the Observed 

Splitting fibers (Dunn) Difference Percent. 
Frog B 
Thigh r. 2577 2623 — 46 —1.7% 
Thigh 1. 2608 2600 + 8 +0.3% 
Frog C 
Thigh r. 2884 2814 + 70 +2.5 
Thigh I. 2887 2777 +110 +4.0 


If we take the differences between the estimated and ob- 
served number of fibers to the thigh in all three frogs, B, C and 
B II, we find that the total is 251 fibers, neglecting signs, or 
deducting the — fibers from the + fibers, an excess of + 189 
fibers. In the first case the difference amounts to an average 
of 1.4%, neglecting signs, or +1.1% when the signs are taken 
into account. 

This is a very satisfactory approximation of the estimate 
to the observed number. 

In the case of the shank we have only the two observa- 
tions on Frog B II, with which to make the test. Here the 
approximation is. closer, being 16 fibers when the signs are neg- 
lected, or a difference of — 12 fibers regarding the signs. This 
gives the percentage values of .37% and—.28%. In the case 
of the foot, owing to the fact that we have not the data for con- 
trolling the splitting fibers, the divisions of which have a simi- 
lar distribution, but credit the differences between the estimated 
and observed numbers to this class, the test of the approxima- 
tion cannot be made, as it has been for the thigh and shank. 
However it will be well to point out the reasons in detail for 
thinking that in the nerves to the foot, this class of fibers will 
account for the differences which have been found. 

In the first place, Dr. DuNN (1902, p. 315, Table XII) has 
shown in Frog B II, between S:, just below the branches to the 
thigh and T and P just at the entrance of the trunk into the 
shank, an interval from which no branches are given off—, that 
at the lowest level on the right side there was a gain of 210 
fibers and on the left, 184, or 5.3% on the right and 4.5% on 
the left, the percentage in each case being based on the number 


Donatpson, Law of [nnervation. 251 


at S: This is proof positive of the splitting of the fibers 
within the nerve trunks. Moreover, in our calculations which 
were made to get the estimated number of fibers going to the 
thigh, the average value of the splitting fibers of class ‘‘a”’ 
in these case was 2.9%. For the two shanks of Frog B II it 
was 9.2%, while that demanded for the foot was 9.9%, all 
these percentages being based on the calculated number of 
unbranched fibers to each segment. 

The significance of the foregoing paragraph is the follow- 
ing: When we assume the splitting fibers of this class to this 
amount, the results of the estimation and observation agree 
closely, and that in the shank where good agreement is ob- 
tained, we have assumed 9.2% of these splitting fibers, where- 
as in the foot, at which level our results cannot be tested in 
the same way, but where we should expect a prioria somewhat 
greater number of splitting fibers of this class, we are com- 
pelled to assume only a slightly greater proportion, namely 
9.9 0. 

The claim made, therefore, for the splitting fibers at this 
level of the foot has ample indirect evidence in its favor. 

Just at this point it may be fitting to call attention to the 
fact that the law here enunciated is deduced from the results 
already presented in four different investigations: HarpEsty 
(1899), DonaLpson and SCHOEMAKER (1900), and Dunn (1900 
and 1902), that these were all independent and undertaken 
without any thought of the present problem, and that the data 
taken from them have been here employed without any modifi- 
cation whatsoever. The only special investigation taken up 
after this problem was formulated was the study of the areas of 
skin in the several segments of the leg, and the observation by 
Dr. Dunn on the nerves in Frog C II, with a view to testing 
whether the results on Frog B II might be considered as gen- 
erally applicable. 

This examination of the data from which the deduction 
has been made goes a long way to protect it from the criticism 
of bias, either unconscious or otherwise. 

It has then been possible to show that, allowing fo 


252 JouRNAL OF COMPARATIVE NEUROLOGY. 


splitting fibers, the number going to any segment of the leg, 
calculated on the basis of the proportional area of skin and the 
proportional weight of the muscles, agrees very closely with 
the number observed. . 

To generalize these results, it is desirable to know, not 
only what percentage of all the fibers which enter the leg goes to 
any segment, as given by the formulain Table X, but also the 
percentage of that number that should be added to the supply 
of a given segment, to make it equal to the observed number. 

This can be obtained by tabulating in each case the esti- 
mated and observed numbers, finding the number of fibers by 
which they differ and the percentage value of this number on 
the basis of the number estimated. 

The averages of the percentages in each case then show 
the proportion of fibers which should be added to the estimate 
in order to represent the number which would probably be 


observed. 
TABLE XXV. 
Showing the percentage of the estimated number of fibers to be added 
in order to make the estimate agree with the number observed. 


Thigh 
Estimate Added Required 
Frog j 46% in calculation to balance 
Ber: 2426 151 = 6.2% 197 
13}, Ue 2477 131 = 5.3% 138 
Cox: 2765 i 4-3% 49 
Ci 2754 133 = 4.8% 23 
Beir. 3269 237 = 7.2% 239 
BIL 3279 228 = 7.0% 212 
Total 16970 999 59% 858 = 5.0% 
Difference = .9% 
Shank 
25-3% 
Ballo 1796 320 = 17.8% 334 
Balle 1802 308 = 17.1% 306 
Total 3598 628 = 17.5% 640 = 18% 
Difference = 0.5% 
Foot 
28.7% 
BIl. r 2042 455 = 22.3% 455 
Baal 2048 436 = 21.3% 436 
Total 4090 891 = 21.8% 891 = 21.8% 


Difference = 0% 


Donatpson, Law of Lunervation. 253 


A glance at Table XXV shows (1) that to make the 46% 
of the fibers entering the leg equal the number that will prob- 
ably be observed to the thigh, 5% of itself must be added to it. 

(2) That to make the 25.3% of the fibers entering the leg 
equal the number that will probably be observed to the shank, 
18% of itself must be added to it. 

(3) That to make the 28.7% of the fibers entering the leg 
equal the number that will probably be observed to the foot, 
21.897 of itself must be added to it. 

By the aid of these results, therefore, if one is given the 
number of fibers entering the leg, it is possible to estimate 
with a high degree of accuracy, the number which will be 
found going to any segment. 

(5) The number of medullated fibers distributed as muscular 
and cutaneous nerves to cach segment of the leg. 

The efferent nerve fibers in a given segment are present in 
numbers proportional to the weight of muscle and the afferent 
according to the area of skin. The muscular nerves must 
contain all the efferent or motor medullated fibers, but it would 
certainly be very unexpected to find that the cutaneous nerves 
contained all the afferent fibers. . 

The number of cutaneous and muscular fibers has been 
determined in every instance for both thigh and shank by Dr. 
Dounvy, and with this we can compare the distribution, calculated 
on the assumption that all the afferent fibers have a cutaneous 
distribution. 

In the following table, in each instance, the calculated 
sensory (afferent supply, Cal. S.), is compared with the ob- 
served cutaneous (Obs. Cut.) supply—the calculated number 
always being the /arger. The absolute difference in number of 
fibers is noted and its percentage value given—the number 
forming the observed cutaneous supply, being taken as a stand- 
ard in calculating the percentage. In like manner, the 
observed muscular (Obs. Mus.) supply is compared with the 
calculated motor (Cal. Mot.), the observed muscular supply 


always being the /arger—the absolute difference in number of 


254 JOURNAL OF COMPARATIVE NEUROLOGY. 


fibers as well as their percentage values are also given, the 

observed number being taken as the standard. 

The shank is tested in the same way. For the foot 
observations are lacking. A study of the table will show, 
_first, that the afferent or sensory fibers going to the muscles of 
the thigh are from 45% to 3.8% of the number of the cutaneous 
nerve fibers, while the excess of the muscular nerves of the 
thigh over the calculated number of motor fibers, is from 
25.1% to 3.1% of the muscular nerve fibers. 

In the case of the shank, the calculations for afferent 
fibers going to the muscles gives from 52% to 37% and an 
excess of fibers in the muscular nerves of from 34.3% 
to 38.6%. 

TABLE XXVI. 

Comparing the calculated number of sensory fibers with the number 
observed in the cutaneous supply and the calculated number of 
motor fibers compared with the number observed in the muscular 
supply and giving in each case the number of fibers by which 
they differ as well as the percentage value of this number. ‘The 
observed numbers being used as the standards. 


Difference 


Cutaneous Muscular 

Frog Thigh Abs. I Abs. % 
Bar: Vales 1281 Obs. Cut. 876 405 = —45.0% 

Obs. Mus. 1747 Cal. Mot. 1296 "451 = 425.1% 
Beale Calais: 1297. Obs. Cut. 871 426 = —48.94% 

Obs. Mus. 1729 Cal. Mot. 1311 418 = + 24.24 
Gan: Calis: 1436 Obs. Cut. 1279 217 = —17.84% 

Obs. Mus. 1535 Cal. Mot. 1451 107 = + 6.8%, 
Gaal CalsS: 1424 Obs. Cut. 1219 145 = —I1.3% 

Obs. Mus. 1558 Cal. Mot. 1450 85 =+ 5-5% 
Beklon Galles: 1743) VObs. Gnte1675) | (65— 3.8Y, 

Obs. Mus. 1830 Cal. Mot. 1763 67 = + 3.6% 
$2 1) ae Gales: 1748 Obs. Cut. 1676 72 = — 4.39% 

Obs. Mus. 1805 Cal. Mot. 1749 56 = + 3.1% 

Shank 
bid UG ee Cal: S- 1376 Obs. Cut. 1003 373 = —37.09% 

Obs. Mus. 1127. Cal. Mot. 740 "387 = ++34-3% 
BeWIGle Cals: 1371 Obs. Cut. go3 468 = —52.07, 


Obs. Mus. 1205 Cal. Mot. 739 466 = + 38.6% 


DonaLpson, Law of Lunervation. 255 


From these data it appears probable that in all these frogs 
and in every segment of the leg, some afferent fibers are dis- 
tributed, with the muscular nerves. Moreover, the proportion 
thus distributed is shown to be very variable in the case of 
the thigh. 

On comparing the proportion of these afferent fibers going 
to the muscles in the case of the shank, we find that the num- 
ber is relatively larger than to the thigh. The comparison in 
the case of the shank B II, should be made, not with the data 
for the thighs of all the frogs, but with the data for the thighs 
of frog B II. Thus, in round numbers, there are in this 
instance, 4% of these afferent fibers to the thigh and 45% to 
the shank. 

The better supply of the muscles with afferent fibers as we 
pass to the more distal segments of the limb, is what we 
should expect a priori, though in this case the difference 
appears very large. 

Conclusions. 


The present study is concerned with the number of med- 
ullated nerve fibers going to the different segmengs of the 
frog’s leg. 

1. The nerve fibers entering the leg being considered as 
so many separate lines of connection with the several segments 
are found to. be distributed in accordance with the law that the 
efferent fibers are present in proportion to the weight of the 
muscle, and the afferent in proportion to the area of skin. 

2. When this statement is reduced to numerical terms, it 
is expressed by the following formula: 


Of the fibers entering the leg of the frog, 


46.07%, go to the skin and muscles of the Thigh. 
25.3% go to the skin and muscles of the Shank. 
28.79, go to the skin and muscles of the Foot. 


3. Since some of the fibers split after entering the leg, 
the numbers found in the nerve branches to the segments are 
larger than the number of single pathways assigned to the seg- 
ment by the formula. Calculation shows that to determine the 
number which will probably be observed in each case, the 


256 JouRNAL OF COMPARATIVE NEUROLOGY. 


number given by the formula must be increased by a certain 
percentage of itself: 


For the Thigh by 5.0% 
For the Shank by 18.0% 
For the Foot by 21.8% 
4. When due allowance is made for the splitting fibers, 
it is found that the agreement between the estimated and 


observed numbers is close: 


Difference. 
oye NSN NGS average of 6 cases = +1.10% 
Rorthershankse2 oe asee= average of 2 cases = — .28¢ 


ForathesHootees=—— pee average of 2 cases = 


*The necessary observations for the foot are lacking, though the agree- 
ment would probably be close there also. 


5. Some of the afferent fibers are distributed to the 
muscles. The proportion in the thigh is highly variable. The 
one case available shows a larger proportion of afferent fibers 
to the muscles of the shank than to those of the thigh, sug- 
vesting that in the foot the proportion would be still greater. 

Thus, while the number of afferent fibers in a segment 
appears to be in proportion to the area of the skin, yet the 
distribution of these fibers is both to the skin and the muscles. 
The significance of this arrangement can only be determined 
by work on other forms, both higher and lower than the frog, 
in the zoGlogical scale. 


BIBLIOGRAPHY. 


Date, H. H. 

19g0t. Observations chiefly by the Degeneration Method, on possible 
Efferent Fibers in the Dorsal Nerve Roots of the Toad and Frog. 
Journ. of Phystol. (Foster). Vol. XXVII, tgor. 

Donaldson, H. H. 

1898. Observations on the Weight and Length of the Central Nervous 
System and of the Legs, in Bull-frogs of Different Sizes. Journ. 
Comp. Neurol. Vol. VIII, No. 4, Dec. 1898. 

Donaldson H.H. and Schoemaker, D. M. 

1900. Observations on the Weight and Length of the Central Nervous 
System and of the Legs in Frogs of Different Sizes (Rana virescens 
brachycephala, Cope). Journ. Comp. Neurol. Vol. X, No. 1, 
Feb. 1900. 


DonaLpson, Law of Lnnervation. 27 


Donaldson, H. H. 

1902. On the formula for Determining the Weight of the Central Ner- 
vous System of the Frog from the Weight and Length of its Entire 
Body.—The Decennial Publications of the University of Chicago. 
Vol. X, 1902. The Untversity of Chicago Press, Chicago. 

Dunn, E. H. 

1900. The Number and Size ot the Nerve Fibers Innervating the Skin 
and Muscles of the Thigh in the Frog (Rana virescens brachyceph- 
ala, Cope). Journ. Comp. Neurol. Vol. X, No. 2, May 1900. 

Dunn, E. H. 

1902. On the Number and on the Relation between Diameter and Dis- 
tribution of the Nerve Fibers Innervating the Leg of the Frog, 
Rana virescens brachycephala, Cope. /ourn. Comp. Neurol, Vol. 
XII, No. 4, 1902. 

Gaupp, Ernst. 

1897. A. Ecker’s und Rk. Wiedersheim’s Anatomie des Frosches. 

Abtheil. 1, i. 1, pp: 191, 1897, 
Hardesty, I. 

1899. The Number and Arrangement of the Fibers Forming the Spinal 
Nerves of the Frog (Rana virescens). Journ. Comp. Neurol. Vol. 
IX, No. 2, 1899. 

Horton-Smith, R. J. 

1897. On Efferent Fibers in the Posterior Roots of the Frog. /ourn. of 

Physiol. (Foster). Vol. XXI, 1897. 
Wana, Julius. 

1898. Ueber abnormen Verlauf einzelner motorischen Nervenfasern 

im Wurzelgebiet. Arch. f. d. ges Physiol. Bd. LXXI, p. 555, 1898. 


4 
hi 
hy Yi ibe 
Ki 


i 
ven 
iv ? 
iA i iy 
; it ay : 
wt p 
Wi, 
| oar i" 
: a it a 
ms 
ie 
é 


VoutumeE XIII. 1903. NUMBER 4. 


oR ELE 


JournaLt or Comparative Neuro.oey. 


Pie RARE OF THE NERVOUS IMPULSE IN. THE 
VENTRAL NERVE-CORD OF CERTAIN WORMS. 


By O. P. JENKINS AND A. J. CARLSON. 


(From the Hopkins’ Seastde Laboratory and the Physiological Laboratory of 
Leland Stanford, Jr. University.) 


With 14 Figures. 


The physiological properties of the muscles in different 
species of worms have been investigated by BrEDERMAN (1, 
1889), Furst (2, 1889), UrxkitLy (3, 1896), Borrazzi (4, 
1898), SrrauB (5, 1900), and BuppincTon (6, 1902), and 
the physiological effects of sectioning the ventral nerve cord, 
extirpation of portions of the same, and extirpation of the 
cesophageal yanglia have been studied by Logs (7, 1894), 
FRIEDLANDER (8, 1894), and Maxwett (9, 1897); but no 
observations on the rate of propagation of the nervous impulse 
appear to have been made in this phylum. In this paper we 
record some measurements, done by the graphic method, of the 
rate of nervous impulse in the following species: 


Cerebratulus sp. Sthenelais fusca Johnson 

Aulastomum lacustre Eunice sp. 

Cirratulus sp.? Nereis virens Sars =N. branti Ehlers 
Arenicola sp. Nereis sp. 

Bispira polymorpha Johnson Lumbriconereis sp. (a) 

Aphrodite sp. Lumbriconereis sp. (b) 

Polynoe pulchra Johnson Glycera rugosa Johnson 


1 The two nemertians worked on were identified for us by Professor R. W. 
Cok of Yale University as belonging to this genus, species probably new. 


2 The marine annelids were identified for us by Professor H. P. JOHNSON 
of Harvard University. Some of them are new species described by Professor 


260 JOURNAL OF COMPARATIVE NEUROLOGY. 


The work on the marine forms was done at the Hopkins’ 
Seaside Laboratory, Pacific Grove. 

A simple muscle-nerve preparation suitable to the graphic 
method is not obtainable in the worms. The preparation we 
employed consisted of the ventral nerve-cord in connection with 
an anterior or a posterior part of the worm serving as a reacting 
portion. This may be called a muscle-nerve-cord preparation 
to distinguish it from the simple muscle-nerve preparation. 
The length of the nerve-cord obtainable for this preparation is 
considerable in some members of the group, and the muscle 
part, especially if prepared from the anterior end of the animal, 
is strong enough to raise a light lever. The extreme posterior 
end of the worm can be used for a reacting portion in few forms 
only, owing to its greater fragility. The length of the reacting 
portion can be suited to any case; we found a length that gave 
a height of contraction of 10 to 20 mm. as magnified by the 
lever most convenient, but in worms with well developed chit- 
enoid epidermis a less height of contraction had to suffice. . 

The apparatus used in the present work was much the same 
as that used by us in the work on the determination of the rate of 
impulse in molluscs (17, 1903). Many difficulties were en- 
countered in preparing the worms for experimentation, but af- 
ter many unsuccessful attempts the following method was found 
most serviceable. The animal was securely fixed by dissecting 
needles, ventral side down, to the moveable floor of a large 
moist chamber supported on a universal stand, the point of fix- 
ation being sufficiently posterior or anterior to give the desired 
length of the reacting portion. In case it was desired to 
prepare the reacting portion from the anterior part of the ani- 
mal the extreme posterior end was then secured in the same 
manner, care being taken that the anterior and posterior points 
of fixation were so close together that maximal longitudinal 
contraction of the worm between these two points exerted lit- 


JOHNSON (Preliminary Account of the Facific Coast Annelids, Prec. Cal. Acad. 
Sci., 3rd ser. Vol. I, No. 5, 1897; the Polychaeta of the Puget Sound Region, 
Proc. Bost. Soc. Nat. Hist., Vol. 29, No. 18, 1901); and the species not named 
are now in his hands for study. 


JENKINS AND CARLSON, WVervous Impulse in Worms. 261 


tle or no pull on the needles, for when this is the case the worm 
breaks in two. The body of the worm was secured to the 
board as before in two places: one about I to 2 cm. anterior to 
the last and the other about the same distance posterior to 
the first point of fixation, care being taken in all cases not to 
injure the nerve cord. Between the two anterior and posterior 
fixed points the body of the animal was laid open by a longitud- 
inal dorso-median incision through the body wall. The body 
wall was pinned out to either side and the exposed viscera were 
turned to one side or removed. The nerve cord was dissected 
free from the adjacent tissue and the entire musculature com- 
pletely severed near the reacting portion, so that this was in con- 
nection with the rest of the body by the nerve cord only. The 
freed portions of the nerve cord were placed on the distal and 
the proximal pair of platinum electrodes respectively, and the 
muscle was connected with the lever by means of a hook and 
thread passing over a friction wheel at the end of the platform. 
This general account applies more especially to forms like the 
Glyceride and the Lumbriconereidz, but numerous devices 
such as narcotics, decapitation, etc., had to be resorted to to meet 
the exigencies in other groups. Variations from the general 
method are noted with each group. 

The complete freeing of the nerve cord from adjacent tis- 
sues without impairing its function is extremely difficult, if not 
practically impossible, in most worms, partly because of the 
small calibre of the cord, and partly because of its more or less 
complete investment by the ventral musculature. In Nereis virens 
thenerve cord is largeand therefore comparatively easily freed ; in 
Aulastomum, Aphrodite and, to some extent, in Arenicola, the 
cord runs free in the body cavity; but with these exceptions 
we never succeeded in completely freeing the cord from the 
ventral musculature. 

The severing of the dorso-lateral musculature (the ventral 
musculature is severed by the dissecting out of the nerve cord) 
between the two points of fixation nearest the reacting portion 
was done to prevent complications by possible myelo-conduction. 
At the time, we were not acquainted with FRIEDLANDER’S obser- 


262 JouRNAL OF COMPARATIVE NEUROLOGY. 


vation (8, 1894) on Lumbricus that the contraction of the worm 
on stimulating any part of the body does not extend beyond 
the segment where the ventral nerve cord has been severed. 
But this severing of the musculature involves considerable in- 
jury, especially to worms like the Sabellide and Aphrodite, 
and to avoid this we ascertained by severing the nerve cord and 
stimulating the worms anterior and posterior to the cut that 
the contraction with which we had to do in our experiments 
never passes beyond the segment where the cord is cut, either 
in the postero-anterior or the antero-posterior direction. This 
fact makes the operation unnecessary, and it was therefore dis- 
pensed with in the subsequent work. 

Owing to the necessity of having the fixation points, that 
is, the points of application of the two pairs of electrodes, close 
enough together to prevent the preparation from breaking off 
when being stimulated, the actual distance between the elec- 
trodes could not be taken as a measure of the distance of nerve 
cord traversed by the impulse. At the end of the experimen- 
tation the segments were counted and the distance measured, or 
more properly estimated, the preparation being stretched as 
nearly as possible like the normally crawling worm or after it 
was killed in fresh water. The last method has this advantage 
that it gives uniform results but it may not approximate 
the true distance any closer. It goes without saying that 
neither method gives accuracy. This is especially true of 
forms with only slightly developed chitenoid epidermis like the 
Sabellidz and Glyceride and therefore capable of great elonga- 
tion and contraction. In the Nereide the distance can be mea- 
sured more accurately. In Aphrodite and Aulastomum the 
entire length of the cord was dissected out and measured. In 
addition to this difficulty of obtaining accurate measurements of 
the nerve cord other difficulties and serious sources of error 
were encountered in the experimentation, chief among these 
the activity or ‘‘spontaneous”’ contractions of the reacting por- 
tion and its unequal degree of relaxation at the moment the rec- 
ords were obtained, and the fact that the same strength of stim- 
ulus usually caused greater contraction when applied by the prox- 


JENKINS AND Cartson, Nervous Lmpulse in Worms. 263 


imal electrodes to the nerve cord near the reacting musculature 
than when applied by the distal electrodes some centimeters 
away from the reacting portion. The incessant relaxation and 
contraction caused, no doubt, in part by the injury in the 
dissection and by the pull of the lever, are almost entirely ab- 
sent in forms like Bispira and Cirratulus; they are quite pro- 
nounced in Glycera and Lumbriconereis and even more so in 
Aulastomum, while in Lumbricus they present difficulties 
which we have thus far been unable to surmount. The unequal 
relaxation of the musculature at the moment of stimulation 
caused variations in the height of contraction, which, together 
with the unequal efficiency of the distal and proximal stimuli, 
rendered comparison of the tracings very difficult, as little 
guidance could be had in the height of contraction. We found, 
however, that in many instances the height of contraction varied 
50% without any appreciable variation of the latent time. But 
the latent time seemed to vary with the strength of the stimu- 
lus, and stronger stimuli were always used for the stimulation 
by the distal electrodes to secure approximately the same 
height of contraction as on proximal stimulation. The elec- 
trodes nearest the reacting musculature were always placed one 
to four centimeters from it and additional precautions were taken 
to insure against direct stimulation of the muscle by escape cur- 
rents. 

The measurements are tabulated in summaries for each 
species, but to give an idea of the make-up of these summaries 
the standard deviation and the coefficient of variability of indi- 
vidual measurements of one typical experiment are given. The 
figures present a representative pair of records from each 
species. The relaxation phase of the myograms is much pro- 
longed and in many cases very irregular, but as this does not 
directly concern the present work only the part of the tracings 
showing the latent time and the height of contraction is given. 

Most of the experiments on the marine forms were carried 
out during the summer of Ig01 and 1902, but some records 
were also obtained during the winter seasons. In the winter 
the temperature of the aquarium varied from 11°C. to 20°C., 


264 JOURNAL OF COMPARATIVE NEUROLOGY. 


in the summer from 12°C. to 14°C. The temperature recorded 
in the case of Aulastomum is that of the room. 


Cerebratulus sp. 


Several species of Nemertians are found in abundance at 
Pacific Grove, but most of them are of too delicate a structure 
to be experimented with by the method employed by us. 


a N 


a nn A eee eee 


POTTY TRICE et IT ILC UML DUA WALA T LAC ASA SSA asm 


WAVY 


Fig. 1. Cerebratulus sp. Postero-anterior. Length of nerve cord between dis- 
tal and proximal electrodes: 5 cm. Rate: 8cm. persec. Time: 50d. 
Vv. per sec. 


We succeeded in securing but two specimens of Cerebratu- 
lus of sufficient strength. These specimens were about 30 cm. 
in length, but showed the same tendency to break into small 
fragments on irritation, that was manifested by their smaller 
relatives. Decapitation failed to quiet them. However, we 
secured an unbroken portion of a few centimeters in length from 
the anterior part of each specimen. The Nemertians possess 
one dorsal and two lateral longitudinal nerve cords, in which 
they resemble the Platodes rather than the Annelids. No at- 
tempt was made to isolate the nerve cords. The stimulus 
reached the cord by application of the electrodes to the ventral 
side of the body. Single induction shocks were not always 
sufficient for excitation from the distal point of stimulation. 
The required intensity of the stimulus is much greater in this 
species than in any of the Annelids included in this work. 

The length of the portion of the nerve cord between the 
two points of stimulation was determined in the specimen while 
crawling in the aquarium. 


JENKINS AND Carson, Nervous lmpulse in Worms. 265 


EXPERIMENT No. 1, Table I, Dec. 26, 19QI. 


Distal Proximal 
No. of records 6 6 
Mean iatent time 1.76 sec. 0.29 Sec. 
Standard deviation 0.174 sec.| 0.06 sec. 
Coefficient of variability 10 .20 
Length of cord: 8cm. Rate: 5.44 cm. per sec. 


TABLE I. 


Summary of experiments on Cerebratulus sp. _Postero-anterior rate. 


No. of No. of records Length of | Ry 
experiment Distal Proximal cord in cm. ateia eu. 
: 2 8 5-44 
= 5 5 5 | 9.0 


Aulastomum lacustre. 
(Horse Leech.) 


Through the kindness of Mr. J. C. Brown of the Depart- 
ment of Animal Biology of the University of Minnesota, we re- 
ceived a number of large individuals of this species from Lake 
Vermillon, Minn. They thrived well in the aquarium, some 


kG 


AVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAVAN 


Fig. 2. Aulastomum lacustre, Postero-anterior. Length of nerve cord between 
distal and proximal electrodes: 11cm. Rate: 63.8 cm. persec. Time: 


50 d.v. per sec. 


being kept for over three months. The preparations used may 
be therefore regarded as from individuals in good condition. 


266 JOURNAL OF COMPARATIVE NEUROLOGY. 


The muscle-nerve-cord preparation was made as in the other 
members of this series. It was found after experimenting with 
one specimen that the dissection of the cord for its whole length 
produced too great injury to it, and such extended dissection is. 
unnecessary. The cord is fairly free and easily accessible to 
the electrodes. No precaution against its breaking itself in 
pieces is necessary with the leech. Unfortunately, the dis- 
section necessary to make the preparation caused the posterior 
portion of the animal to contract so strongly that it would not 
react to further stimulus and hence could not be used to make 
records of contractions. Consequently we were unable to de- 
termine the antero-posterior rate in the cord. The head end, 
however, being free from this prolonged contraction allowed 
the determination of the postero-anterior rate. This was not 
without difficulties, for the almost incessant motion of the an- 
terior part of the animal, thus continually displacing the lever 
attached to it, makes it necessary to watch very closely to catch 
the moment of rest of the animal, coinciding with proper ad- 
justment of apparatus. Still, when all was right, the stimulus 
gave prompt and decisive reaction. These conditions made it 
obviously impossible to take the records at equal intervals. 

It was found necessary to make use of the interrupted cur- 
rent, for, while single induction shocks were efficient as a stim- 
ulus when applied at the point of the cord nearest the muscula- 
ture used, they failed to produce a definite contraction when ap- 
plied at the distal point. This was true for both directions in 
the cord. However, a single induced shock applied to a point 
in the posterior end of the cord generally produced the indefi- 
nite motion of swimming or crawling. 

At the conclusion of each experiment the worm was killed 
in 5% alcohol, the cord dissected out, and measured after tak- 
ing the precaution to straighten out the loops formed in the in- 
terganglionic portions produced by the contraction of the body 
of fhe leech 

The temperature of the room varied from 16°C, to 21°C. 


EXPERIMENT No. 2, Table II, Oct. 5, Igo1. 


Distal Proximal 
No. of records 23 23 
Mean latent time 0.28 sec.| O.II sec. 
Standard deviation 0.03 sec.| 0.004 sec. 
Coefficient of variability .10 |  .036 


Length of cord: 10.25 cm. Rate: 59.4 cm. per sec. 


JENKINS AND Carson, Nervous Impulse in Worms. 267 


TABLE II. 


Summary of experiments on Aulastomum lacustre. 
Postero-anterior rate. 


No. of No. of records Length of } ; 
experiment Distal Proximal cord in cm. Rate in em.; 
, ae i ah 10.50 42.0 
2 23 23 10.25 59.4 
3 i + 10.50 39.9 
4 28 27 10.50 57-7 
5 15 18 10.25 48.2 
6 8 9 10.50 45.1 
7 29 25 11.00 47.0 
8 18 21 10.50 87-1 
9 II 15 II.00 572 
10 12 6 10.50 80.8 
Il 26 16 II.00 91.3 
12 16 19 11.00 Bye 
13 18 26 10.50 52.5 


Mean rate: 56cm. per sec. 
Standard deviation: 16. 
Coefficient of variability: ,28. 


Cirratulus sp. 


This species is very common at Pacific Grove where it is 
found under the rocks and in their crevices throughout the lit- 
toral zone. It reaches a length of fifteen to twenty cm. The 
posterior portion of the body is very slender and fragile and too 


DIDIPIPPPPPPPLPLIIIIPIPUIPPIPLIISPPLPPIPIPPLSPSPPSSPLPPP PPP IIPLPP PLP PPS IIIS PLP IP LL SL LI 


Fig. 3. Cirratulus sp. Postero-anterior. Length of nerve cord between distal 
and proximal electrodes: 9 cm. Rate: r00cm. persec. Time: 50 d.v. 
per sec. 


feeble to raise the lever of the apparatus, consequently the an- 
tero-posterior rate was not determined. The head end, how- 


268 JOURNAL OF COMPARATIVE NEUROLOGY. 


ever, reacts well to stimuli to different points in the cord and 
allows the determination of a postero-anterior rate. On first 
making the preparation the head end contracts strongly, but in a 
few minutes it relaxes, and afterwards the preparation is fairly 
quiet, reacting only to the stimulus from the electrodes. A 
single induced shock applied to a point in the posterior portion 
of the cord usually produced contractions in its immediate vi- 
cinity which extended a short distance only, not reaching the 
head. However, a weak interrupted current of short duration 
produced a contraction which reached the anterior end. 

The preparations fatigue quickly, so that only a few com- 
parable tracings can be obtained in each case. 

The length of the nerve cord was measured in the prepara- 
tions after being killed in fresh water. 


EXPERIMENT No. 4, Table [II, postero-anterior, Aug. 16, 1901. 


Distal Proximal 
No. of records 3 3 
Mean latent time 0.28 Sec. 0.17 sec 
Standard deviation 0.009 sec. 0.01 sec. 
Coefficient of variability .03 .05 
Length of cord 10 cm. (105 segments). Rate: 90.9 cm. per sec. 


TABLE III. 
Summary of experiments on Czrratulus sp. 


Postero-anterior rate. 


No. of No. of records Length of : 

. = . Rate in cm. 
experiment Distal | Proximal cord in cm. 

I 3 | 3 13 85.8 

2 6 | 8 15 82.5 

3 4 4 18 86.4 

4 3 3 10 90:9 

5 2 5 9 99.9 

6 4 4 9.5 86.4 


Mean rate: 90 cm. per sec. 
Standard deviation: 1.66. 
Coefficient of variability: .o13. 


Arenicola sp. 


This worm as judged by the number of its egg masses seen 
during the breeding season is common in the vicinity of Pacific 
Grove. However, the sand in which it lives is so filled with 
large rocks that it is difficult to collect it. Only three work- 


able specimens were obtained. It is very sluggish in its move-. 


Ee 


JENKINS AND CarRLson, Nervous Impulse in Worms. 269 


ments. In attempting to make the nerve-cord-muscle prepara- 
tion the musculature is thrown into strong and prolonged con- 
traction. That of the circular musculature was the most marked 
especially at the points where the worm was fastened to the 
floor of the apparatus. This contraction of the circular muscu- 
lature was so vigorous as to prevent the longitudinal muscula- 
ture from shortening the body and thus acting on the lever. A 


—_—________ NWN 


LLLLIILISIGLIPIMISPISIIISSID IIS SDI IIIS SINIIISISILPIS SII 


Fig. 4. Arenicola sp. Antero-posterior. Length of nerve cord between distal 
and proximal electrodes: 10cm. Rate: 125 cm. persec. Time: 50d.v. 
per sec. 


dorsal longitudinal slit throughout the preparation did not les- 
sen the effect of the circular contraction, but put the prepara- 
tion in such a condition that the contraction of the longitudinal 
layer on stimulation of the cord bent the preparation to the in- 
ner side and in this way gave opportunity to attach the record- 
ing lever. The nerve cord is about 0.7 mm. in diameter and 
easily separated from the adjoining tissue. The interrupted 
current was used, since single induction shocks proved inefh- 
cient. 


EXPERIMENT NO. 3, Table IV, antero-posterior, temp. 14.5°C, July 9, 1902. 


Distal Proximal 
No. of records 12 7 
Mean latent time 0.20 sec. 0.12 sec. 
Standard deviation O.OL sec. 0.007 sec. 
Coefficient of variability 205 .06 


Length of cord: tocm. Rate: 126 cm. per sec. 


270 JOURNAL OF COMPARATIVE NEUROLOGY. 


TABLE IV. 
Summary of experiments on Avenicola sp. 


Antero-posterior rate. 


No. of No. of records Length of : 
: Rance : Rate in cm. 
experiment Distal Proximal cord in cm. 
I 26 14 9 128.7 
2 41 30 15 106.5 
3 12 7 10 126.0 


Mean rate: 120.6 cm. per sec. 
Standard deviation: 9. 
Coefficient of variability: .07. 


Bispira polymorpha. 


This large tube inhabiting worm is very abundant near the 
lower limits of the tides. Their tubes are so well hidden in the 
crevices of the rocks and so firmly attached to them, that it is 
only with difficulty and great care that they can be removed 
without injury to the occupants. When undisturbed the head 


Fig. 5. Bispira polymorpha. Postero-anterior. Length of nerve cord be- 
tween distal and proximal electrodes: 28 cm. Rate: 700 cm. per sec. 
Time: 100 d.v. per sec. 


of the worm with the mass of constantly moving tentacles is pro- 
truded from the tube. On being disturbed it suddenly retracts 
its head and tentacles into the tube. Conformably with its 
mode of life, the musculature of the anterior portion of the 
body is greatly developed, especially that of the longitudinal 


JENKINS AND Car son, (Vervous Impulse in Worms. = 271 


layer, while that of the posterior portion is comparatively little 
developed and is feeble in its action. On the other hand, the 
contraction of the anterior portion was quick and powerful, 
readily lifting a weight of 1000 grams. Consequently while 
the preparations allowed the determination of the postero-an- 
terior rate in the cord, the antero-posterior rate was not ob- 
tained. 

On account of the difficulties in dissecting out the cord 
and the chances of injury to it on approaching it through the 
dorsal side, the nerve-cord preparation in this worm could not 
be made in the same way as was done in the other worms. It 
was treated as follows: The worm was pinned to the board dor- 
sal side down; a longitudinal incision was made through the 
body wall on each side about one mm. from the median line on 
the vental side; this ventral strip was further freed from all con- 
nection with the organs in the body cavity. This strip con- 
taining the uninjured ventral cord was raised and placed on the 
electrodes. While this method, used with care, kept the cord 
uninjured, it presented the apparent disadvantage of placing a 
comparatively large thickness of tissue between the cords and 
the electrodes. This did not, however, prove of any disadvan- 
tage, as vigorous responses were obtained from single break in- 
duction shocks even on stimulation at the posterior point, when 
the secondary coil was 24 cm. fromthe primary. The anterior 
12 to 15 segments were used as the reacting portion, leaving 
from 17 to 20 cm. of nerve-cord between the points of stimulation. 
The lever was weighted with from 20 to 25 grams. Ai single 
break induction shock produces a single prolonged contraction. 
The preparation is not easily fatigued. Anaesthetics were 
found not to be necessary. The length of the nerve used was 
measured after the preparation was killed in fresh water. Their 
length, thus killed, corresponds to that of the animals undisturbed 
in the aquarium but freed from their tubes. In this condition the 
worm is from 12 to 15 cm. shorter than than in its tube. 


EXPERIMENT No. 10, Table V, postero-anterior, temp. 14°C, Jan. 2, 1902. 


Distal Proximal 
No. of records II 7] 
Mean latent time 0.057 sec. 0.024 sec. 
Standard deviation 0.0022 sec. 0.002 sec. 
Coefficient of variability -04 .08 


Length of cord: 20 cm. (150 segments). Rate: 606 cm. per sec. 


272 JOURNAL OF COMPARATIVE NEUROLOGY. 


TABLE V. 


Summary of experiments on Bispira polymorpha. 


Postero-anterior rate. 


No of No. of records Length of ; 
: SSS - Rate in cm. 
experiment Distal Proximal cord in cm. 
I 12 12 18 691 
= x 2 15 555 
3 2 2 18 817 
4 15 16 20 | 952 
5 16 15 19 729 
6 11 II 17 629 
7 8 8 20 714 
8 4 2 20 | 526 
9 6 4 17 680 
10 1I 7 20 606 


Mean rate: 694 cm. per sec. 
Standard deviation: 93. 
Coefficient of variability: .13. 


Aphrodite sp. 


An undetermined species, probably new, of this interesting 
Polychete occurs in the vicinity of Pacific Grove. As it lives 
in rather deep water and seems not to be common it is only oc- 
casionally brought in by the fishermen. We have secured thus 


- ne. 


eee ee eee 


a en mn Nee | ee 
WU AAAAAAA 


Fig. 6. Aphrodite sp. Antero-posterior. Length of nerve cord between distal 
and proximal electrodes: 8cm. Rate: 52.8cm. persec. Time: 50 d.v. 
per sec. 


far but two uninjured specimens, the largest of which measured 
12cm. in length and 5 cm. in breadth. This worm is very 
slow in its movements and it was found to possess a rather low 
irritability since single induction shocks failed to procure a re- 
action when applied to the distal point of stimulation. The 


JENKINS AND Car son, Nervous Impulse in Worms. 273 


complicated dorsal structures of this worm together with the 
powerful contractions on irritation contribute to render the 
making of nerve-cord-muscle preparations difficult. The red 
pigmented nerve cord is nearly free in the body cavity and 
when the body is contracted the cord is thrown into loops, 
which necessitated the taking the cord out and straightening it 
before measuring the length involved. The musculature of the 
posterior end was taken as the reacting portion, and the antero- 
posterior rate only was determined. Although single induction 
shocks were not efficient at the distal points of stimulation to 
produce contractions in the longitudinal muscles of the reacting 
portion, they were found to be sufficient to set up a progressive 
movement of the setz from the point of stimulation posteriorly, 
indicating that their neuro-muscular apparatus is more irritable. 
This seemed to be further indicated by the fact that the animal 
could be quite roughly handled without producing contractions 
of the longitudinal muscles, while the sete reacted to the 
treatment. 


F.XPERIMENT No. 2, Table VI, antero-posterior, temp. 15°C, Jan. 19, 1902. 


‘Distal Proximal 
No. of records 25 18 
Mean latent time 0.422 sec 0.26 sec. 
Standard deviation 0.045 sec. 0.03 sec. 
Coefficient of variability _ +10 SDE 


Length of cord: 9 cm. (17 segments). . Rate; tees 25 cm. per sec. 
ABER Vile 


Summary of Experiments on Aphrodite sp. 


Antero-posterior rate. 


“No of. | po lawee a vecords. ; | Length ae e 
experiment | Distal | Promina | cord in cm. ate In cm. 
I 34 3 } 52.8 
. Sade ee Ce, Nea ee ee 


Average r: rate: 54-5 cm, per sec. 


Polynoe pulchra. 


This worm lives on the surface of the oral tentacles of the 
large sea cucumber, folothuria californica, and in the mouth 
cavity of the large key-hole limpet, Lacopzna crenulata, both of 
which are common at Pacific Grove. This species of Polynoe 
reaches a length of from 4 to5 cm. On account of the delicacy 
of its structure the nerve-cord is exposed with some difficulty. 
The worm also breaks in pieces very readily, neither chloralhy- 


274 JOURNAL OF COMPARATIVE NEUROLOGY. 


drate nor decapitation serving to quiet it. These obstacles 
prevented our obtaining satisfactory records from more than two 
specimens, although many specimens were attempted. The 
preparation fatigues quickly and single induction shocks, unless 
very strong, were not always efficient. Only the postero-an- 
terior rate was determined. Another species of this genus, 
Polynoe brevis, is common at Pacific Grove, but as it reaches 
a length of only 2 or 3 cm. no records could be taken from it. 


wal. BAN sc eae aa re 


a I oe 2 ee ee 


pal SPAS EEN NE PALL) a SRE Oot 2 UE eh ee 


GVA LPALDMPSPSPPISSISSS SSIS ISIN III. 


Fig. 7. Polynoe pulchra. Postero-anterior. Length of nerve cord between 
distal and proximal electrodes: 3.5cm. Rate: 350 cm. per sec. Time: 
50 d.v. per sec. 


EXPERIMENT No. 1, Table VII, postero-anterior, temp. 13°C, Aug. 16, IgoI. 


Distal Proximal 
No. of records 2 2 
Mean latent time 0.127 sec. 0.110 sec. 
Standard deviation 0.004 sec. none 
Coefficient of variability .03 none 


Length of cord: 4 cm. (33 segments). Rate: 230 cm. per sec. 


TABLE VII. 


Summary of experiments on Polynoe pulchra. 


Postero-anterior rate. 


No. of No. of records Length of ’ 
: : : Rate in cm, 
experiment Distal Proximal cord in cm. 
I 2 2 4 236 
2 2 2 3.5 350 


Average rate: 293 cm. per sec. 
Sthenelais fusca. 


This worm seems to be rare at Pacific Grove. Satisfactory 
records for a single specimen only were obtained, and these are 
of the rate of the impulse in the postero-anterior direction. 


JENKINS AND Carson, Nervous Impulse in Worms. — 275 


pea OL Pa Ta) ae Fea Se eS 


LIPOPPOMIALASILIIMIIIPIMSPPIPPLIPLIILIPICELIPIII IPL PL LIAL ILL III 


Fig. 8. Sthenelats fusca. Postero-anterior. Length of nerve cord between 
distal and proximal electrodes: 5 cm. Rate: 250 cm. persec. Time: 100 
d.v. per sec. 


TABLE VIII. 
Detail of experiment on Sthenelats fusca. Dec. 31, 1902. 


Postero-anterior. 


Total latent time in sec. 


Distal Proximal 
.065 -025 
045 .028 
055 .028 
-050 .029 
O45 -028 
.048 .025 
Average: .O51 rch .027 


Transmission time: 0.024 sec. 
Length of cord: 5 cm. (53 segments). 
Rate: 205 cm. per sec. 


Eunice sp. 


This worm, fairly abundant, lives under rocks in tubular 
passages formed by cementing particles of gravel together. It 
attains a considerable length, specimens 25 cm. long being 
taken. It breaks into pieces very readily. For this reason, 
and further, because of its violent struggling in being captured, 
an unbroken specimen was rarely secured. 

When being prepared for experimentation in the usual 
manner, Eunice always broke in pieces unless it was previously 


276 JOURNAL OF COMPARATIVE NEUROLOGY. 


decapitated, but even the decapitated preparation would some- 
times twist itself in two. Chlorhydrate had the same quieting 
effect as decapitation. 

Eunice is very quick in its movements and compares in ir- 
ritability with Polynoe and Nereis. Single induced shocks were 
efficient stimuli, the reacting portion responding generally with 
a single prolonged contraction. 


MIOPMPOLIIWIPPIIIMWIPPPPPPPPPIPIIPOPMIOS_PMVIIVPII PLC IIPIPLIOPPLILIPLLI LI IDID LIL III LIL DL 


Fig. 9. Eunice sp. Postero-anterior. Length of nerve cord between distal 
and proximal electrodes: 9 cm. Rate: 450 cm. persec. Time: 100 d.v. 


per sec. 


The measurements were made after the preparation had 
been killed in fresh water, which gives nearly the same exten- 
sion as when the worm is lying quiet in the aquarium. 

The antero-posterior rate alone was determined. 


EXPERIMENT No. 3, Table IX, postero-anterior, Dec. 26, 1902. 


Distal Proximal 
No. of records 13 12 
Mean latent time 0.050 sec. | 0.038 sec. 
Standard deviation 0.003 sec. | O 0023 sec. 
Coefficient of variability .060 .061 


Length of cord: 5.7 cm. Rate: 474.8 cm. per sec. 


JENKINS AND Cartison, Nervous Impulse in Worms. 277. 


TABLE IX. 
Summary of experiments on Lunice sf. 


Postero-anterior rate. 


No. of No. of records Length of f 
; Se 5 Rate in cm. 
experiment Distal | Proximal cord in cm. 
I 2 | 2 9 473-4 
z 3 3 4.5 409.0 
3 14 | 12 5-7 474.8 
4 13 12 | 13.0 520.0 


Mean rate: 466 cm. per sec. 
Standard deviation: 41. 
Coefficient of variability: .o9. 


Nereis sp. 


This species is very common at this point. It is found 
under rocks and crevices, attaining a length of 15 cm. The 
posterior third of the body is too weak and fragile to raise the 
lever, hence the antero-posterior rate was not determined. Un- 
less it was first decapitated the conditions of the experiment 
invariably threw the worm into violent contractions by which 
it broke into pieces. The activity and irritability of this worm 


ee em tat 


—EEEE—EE—EEEE 


soe ee 
VWAWIIIIJIAIIJIJIJIJIJIJIJIJIJIIIJVIJIIJIJIIWIPIPTIIIJSIIISP IS SINSNINSNINAISSIIS- 


Fig. ro. Nereis sp. Postero-anterior. Length of nerve cord between distal 
and proximal electrodes: 12 cm. Rate: 133.2 cm. persec. Time: 50 
d.v. per sec. 

is much the same as that of Eunice and Glycera. However it 
does not respond to single induction shocks as regularly as 
does Glycera. A single shock applied to the cord posteriorly 
produces a contraction anterior to the point of stimulation, the 
distance to which the contraction extends depending on the in- 
tensity of the stimulus. Still, the contraction does not always 
reach the head even with a high intensity of stimulus. The 
interrupted current is more sure of producing this effect. 


278 JOURNAL OF COMPARATIVE NEUROLOGY. 


The length of nerve-cord between the points of stimulation 
was measured after the preparation was killed in fresh water, in 
which the condition is very near the normal extension of the 
worm. 


EXPERIMENT NO. 2, Table X, postero-anterior, Aug. 7, Igor. 


Distal Proximal 
No. of records 8 9 
Mean latent time 0.095 sec.| 0.061 sec. 
Standard deviation 0.003 sec. | 0.0046 sec. 
Coefficient of variability .031 -075 


Length of cord: 55 cm. (45 segments). 
Rate: 165 cm. per sec. 
TABLE X. 
Summary of experiments on WVerezs sp. 


Postero-anterior rate. 


No. on ie No: of records Length of ’ 

j : ; : Rate in cm. 
experiment Distal Proximal cord in cm. 

1 A 4 4 10.0 164 

2 8 8 5:5 165 

31 9 9 12.0 223-2 

4 3 3 5-0 156.0 

5 4 4 120 | 123.6 


Mean rate: 165 cm. per sec. 
Standard deviation: 32. 
Coefficient of variability: .19. 


Nereis virens Sars (=N. branti Ehlers). 


This species seems to be quite rare at Pacific Grove, since 
but one specimen was secured during extended collecting in 
the summers of 1901 and 1902. The specimen was about 100 
cm. in length and from 1 to 1.5 cm. in diameter, but it unfor- 
tunately broke into several pieces through its contractions when 
being captured and the piece containing the head, which was 
the largest, broke up further in attempting to submit it to the 
experimental conditions. Decapitation served to quiet its mo- 
tions, and, no doubt, a much. longer portion could have been 
obtained if this treatment had been used at first. The ventral 
cord was from I mm. to I.5 mm. in diameter, which large size 
together with its favorable position in the body allowed it to be 
easily dissected free. Two pieces were used, one with the 
head contained 85 segments and gave an available distance of 
cord of 20cm. The other was from near the tail and con- 
tained 25 segments with 6.5 cm. of cord available. These dis- 


JENKINS AND Carson, WVervous Impulse in Worms. 279 


tances were measured while the pieces were crawling in the 
aquarium and represent very nearly the length during the nat- 
ural extension of the worm. 

Single induction shocks applied to the distal point of 
stimulation produced a single prolonged contraction. After 
three or four reactions, single shocks failed at the distal 
point, then the break or make of the direct current was still 
efficient. Since the largest preparation was no more than one- 
fourth the length of the whole worm, it is impossible to say 
whether through the whole length of the cord a single induc- 
tion shock would prove efficient. 


ame en mn PGS 


Dee 


ae INN ee 
SLPLISLPPPPFSSL PS LPGPPLSP SPP SSISLS SPS SPSS IVS SLI ISI III 


Fig. 11. Nere/s virens. Antero-posterior. Length of nerve cord between dis- 
tal and proximal electrodes: 6.5 cm. Rate. 81.2 cm. persec. Time: 56 
d.v. per sec. 


Good reactions were obtained from one portion 24 hours 
after its separation from the rest of the body; in other pieces 
48 hours later only feeble responses were obtained; however 
one piece lived three weeks in the aquarium, when it was de- 
stroyed by an accident. In one piece the rate was taken in the 
antero-posterior direction, in the other postero-anterior direc- 
tion. The rates 89 and 73.4 respectively, differing as they do, 
may not be taken, in the absence of other experiments, as con- 
clusive evidence that the rates in the two directions differ. 


EXPERIMENT No. 1, Table XI, antero-posterior, July 20, 1901. 


Distal | Proximal 
_— - | = - 
No. of records 27 ES 
Mean latent time 0.183 sec. 0.096 sec. 
Standard deviation 0.016 sec. 0.007 sec. 
Coefficient of variability .08 Or 


Length of cord: 6.5 cm. (25 segments). 
Rate: 73.4 cm. per sec. 


280 JoURNAL OF COMPARATIVE NEUROLOGY. 


TABLE Xi. 
Summary of experiments on Nereis virens. 
No. of | No. of records Length of : 
s : Rate in cm. 
experiment Distal | Proximal cord in cm. 
I | 
(antero-posterior) | 27 | Ig 6.5 73-4 
5 | 
(postero-anterior) | 13 | II | 20.0 | 89.0 


Lumbriconereis sp. (a). 


Small specimens of this species of 20 to 30 cm. and from 
2 to 3 mm. in diameter are very abundant in the gravel at the 
lower limits of the tide, but only one large specimen was se- 
cured. This was 75 cm. in length and from 6 to 7 mm. in di- 
ameter, but the specimen was accidentally broken in two pieces 
in being captured. The anterior portion proved of great ser- 
vice since since 150 good records distributed over a period of 
three days were obtained from it. The posterior portion failed 


WIVWVYIIJIIIIIVIYJYAPJIPIPIWIIVIIVIJIJIJI(II{IIIIIIIJPIPAIIPADIIIIIN 


Fig. 12. Lumbriconerets sp. (a) Postero-anterior. Length of nerve cord be- 
tween distal and proximal electrodes: 21cm. Rate: 262.5 cm. per sec. 
Time: 50 d.v. per sec. 

to give equally good results from the fact that in attempting to 
prepare it for experimentation it went into a strongly contracted 
state which continued for three days, although left undisturbed 
in the aquarium. During this time it failed to react to any 
stimuli. This same phenomenon has been observed in some of 
the specimens of Glycera which had been injured. The smaller 
specimens of Lumbriconereis gave so much trouble in this way 
that but few records were obtained from them. These are the 
only free swimming polychetes in which this phenomenon was 
observed. 


JENKINS AND Carson, Nervous lmpulse in Worms. 281 


The nerve-cord-muscle preparation can more easily be 
made from Lumbriconereis than from Glycera on account of its 
squirming less. Like Glycera, it is less easily broken than the 
rest of the series worked with. It is less active than Glycera 
but equally irritable; single induction shocks applied to the cord 
20 cm. from the reacting portion produce good contractions. 
The measurements of the cord were taken after killing the prep- 
aration in fresh water. 


EXPERIMENT No. 1, Table XII, postero-anterior, July 24 to 27, 1goOl. 


Length of cord: 21 cm. (114 segments). 


July 24 First four pairs of records, rate: 300 cm. per sec. 
3-9 p- m, Total No. of records (24 pairs), rate: 65 cm. per sec. 
July 25 First four pairs of records, rate: 221 cm. per sec. 

8-11 a.m. Total No. of records (16 pairs), rate: 65 cm. per sec. 
July 25 First three pairs of records, rate: 210 cm. per sec. 
4-6 p. m. Total No. of records (15 pairs), rate: 94 cm. per sec. 
July 26 First three pairs of records rate: 233 cm. per sec. 
ga.m.-5 p.m. Total No. of records (13 pairs), rate: $0 cm. per sec. 
July 27 First pair of records, rate: 42 cm. per sec. 

ga. m. 


The coefficient of variability of the transmission time of the fourteen pairs 
of records that show a rate above 200 cm. per sec, is .37. 


TABLE XD. 


Summary of experiments on Lumbriconerets sp. (a) 


Postero-anterior rate. 


No. of | No. of records | Length of ; 
- aa = = : Rate in cm. 
experiment Distal Erosimal cord in cm. 
I | 14 14 | 21 241 
2 3 | 3 | 9 45 
eee, 4 4 10 45 


Lumbriconereis sp. (b) 


This worm is very abundant in crevices and under rocks 
where it constructs for itself delicate tubular passages by debris 
and gravel cemented together. It is commonly 30 to 40 cm. 
in length and from 5 to 7 mm. in diameter, but owing to its 
great fragility generally only portions of the worm could be se- 
cured for the preparation. The worm is rather inactive, so 
anesthetics were not necessary. Despite its sluggish habits it 
responds to single shocks almost as readily as Glycera, but the 


282 JoURNAL OF COMPARATIVE NEUROLOGY. 


preparation fatigues much sooner. The measurements of the 
cord were made on preparations killed in fresh water, with the 
exception of preparation No. 7, Table XIII A., which was ex- 
tended during the experimentation to the length given. 


ah a 


Be EEE OOO OeOOEeeeeeeeeeeeesees@9s@s—_” reo 


oT EEE ESSE 


Fig. 13. Lumbriconerets sp. (b) Postero-anterior. Length of nerve cord be- 
tween distal and proximal electrodes: 30cm. Rate: 681 cm. per sec. 
Time: 100 d.v. per sec. 


EXPERIMENT No. 4, Table XIII, postero-anterior, Aug. 7, 1902. 


Distal Proximal 
No. of records 6 6 
Mean latent time 0.060 sec. 0.035 sec. 
Standard deviation 0.004 sec. 0.004 sec. 
Coefficient of variability -O7 at 


Length of cord: 15 cm. (125 segments). ‘Rate: 600 cm. per sec. 


TABLE XIII A. 


Summary of experiments on Lumértconereis sp. (b) 


ustero anterior rate. 


No. of No. of records _ Length of , 
i . Rate in cm. 
experiment Distal Proximal cord in cm. 
I 4 2 20 40 
Z 5 3 13 
3 3 3 16 
4 6 6 | 1) 
5 4 4+ | 12 52 
6 5 5 15 937 
7 9 8 30 768 


JENKINS AND Carison, Nervous Impulse in Worms. 283. 


TABLE XIII B. 


Summary of experiments on Lumbriconerets sp. (b) 


Antero-posterior rate. 


No. of No. of records | Length of ) 
experiment Dista Proximal cord in cm. Rate in cm. 
: a) 4 25 ; 80 
2 3 3 8 160 
3 2 fe) 42 


Glycera rugosa Johnson. 


This worm is fairly abundant in the sands of the lower tide 
limits at Pacific Grove. The largest specimen that we collected 
measured from 35 to 40 cm. in length and about 13 mm. in diam- 
eter, when killed in fresh water. It is very ferocious, even in the: 
aquaria preying on Eunice and Nereis. It is also very active, 


ROR igs FNS a th a ce RRO ae ge A la WS er A Se 
NAA 


fig. 14. Glycera rugosa. Antero-posterior. Length of nerve cord between 
distal and proximal electrodes: 30cm. Rate: 441 cm. per sec. Time: 50 


d.v. per sec. 


making its way through the wet sand with great speed or glid- 
ing through the water by the graceful twistings of its whole 
body. This worm was selected for a beginning in this work 
because of its large size, high irritability and strength. Glycera 
is less fragile than any of other free swimming polychetes 
worked excepting Lumbriconereis sp. (a) which equals it in 
this respect. 


284 JOURNAL OF COMPARATIVE NEUROLOGY. 


In most cases an anaesthetic, chlorhydrate, was necessary to 
quiet the violent squirming in making the preparation. De- 
capitation did not quiet these motions. The preparations re- 
sponded with great regularity to single induction shocks of 
moderate intensity applied to the cord 20 to 30 cm. from the 
contracting portion; when the single shock became inefficient 
the interrupted current of short duration was used. The nerve- 
cord-preparation of this species is not easily fatigued and conse- 
quently allows a great number of records to be taken from each 
specimen. The preparation from one specimen, No. 2, Table 
XIV A, gave good responses 23 hours after it was made. 

It frequently happened that an injury to any part of the 
body caused the musculature, both circular and longitudinal, to 
go into a strongly contracted state, in the immediate region of 
the injury. This condition in some cases continued for 24 
hours, or even more, during which time it is practically impos- 
sible to send an impulse through the cord of the contracted 
portion. In the only case in which an impulse passed through 
so small a distance as 100 segments of the worm a strong inter- 
rupted current was necessary. Even in this case the respond- 
ing contractions were feeble and very much delayed. 


EXPERIMENT No. 2, Table XIV A, postero-anterior, July 16, 1901. 


Distal | Proximal 
No. of records 32 32 
Mean latent time 0.18 sec. 0.10 sec. 
Standard deviation 0.014 sec. 0.003 sec. 
Coefficient of variability .08 .03 


Length of cord: 34 cm. (150 segments). Rate: 425 cm. per sec. 
TABLE XIV A. 
Summary of experiments on Glycera rugosa. 
Antero-posterior rate. 


No. of | No. of records Length of | ; 
. ; | Rate in cm. 
experiment \ Distal Proxima! cord in cm. | 
I | 21 22 34 680 
2 | 33 32 34 | 425 
3 II Il 25 357 
4 22 22 20 ‘400 
5 4 | 4 25 500 
6 | 25 25 17 425 
7 | II 11 20 400 
8 12 Il 30 375 
9 9 9 24 | 271 


Mean rate: 433 cm, per sec. 
Standard deviation: 87. 
Coefficient of variability: .20. 


JENKINS AND Carson, Nervous Impulse in Worms. 285 


TABLE XIV B. 


Summary of experiments on Glycera rugosa. 


Postero-anterior rate. 


No. of No. of records Length of ‘ 
: : Rate in cm. 
experiment Distal | Proximal cord in cm. 
ear ve | i | 7 15 | 420 
2 2 2 18 | 514 
3 | 7 | 7 13 540 
ee NTs i aE 18 | 265 
Mean rate: 435 cm. per sec. 
Standard deviation: 106. 
Coefficient of variability: .24. 
TABLE XV. 
Summary of the rates in the fourteen species worked on. 
ki é _. | Coeff 
Species Direction Race ks cient of 
individuals | in cm. Res 
vari’ bility 
Cerebratulus sp. Postero-anterior 2 5-4-9.0 ah 
Aulastomum lacustre oe oe 13 56.0 .28 
Cirratulus sp. ag ne | 6 | go-0) |) :01g 
Arenicola sp. Antero-posterior 3 eu 2O208s oy 
Bispira polymorpha Postero-anterior IO 6940 | .13 
Aphrodite sp. Antero-posterior 2 54-5 .04 
Polynoe pulchra Postero-anterior 2 293-0 | .19 
Sthenelais fusca ne ye | I 205.0 
Eunice sp. ae < | 4 466.0 .09 
Nereis sp. se ck 5 165.0 .19 
Nereis virens at a I 89.0 | 
Ge oe Antero -posterior I 73-4 
Lubriconereis sp. (a) |Postero anterior 2 45-241 
ef * (Bb) a oe 7 49-937 
ae «« (b) |Antero-posterior 3 42-160 
Glycera rugosa Ob SS 9 433-0 .20 
. ck Postero-anterior 4 435-0 24 


page 268), Bispira (table V, page 272), Eunice (table IX, page 
277), Nereis sp. (table X, page 278), and Glycera (table XIV, 
page 284), the number of individals worked and the relative 
constancy of the individual rates seem to allow the conclusion 
that the mean of the individual rates is representative for their 
respective species. The mean in the groups represented by 
by two or three individuals only is of less value, especially in 
the two species of Lumbriconereis (tables XII and XIII) where 
the individual variations are so great. But in some species 


286 JOURNAL OF COMPARATIVE NEUROLOGY. 


thus scantily represented the individual rates show only slight 
variation, as in Arenicola (table IV, page 270), Aphrodite (table 
VI, page 273), Polynoe and Sthenelais (tables VII and VIII, 
pages 274 and 275), and Nereis virens (table XI, page 280), and 
their representative means may therefore not be far from the 
true rate in these species. 

The rate of the nervous impulse in this series of worms is 
as varied as the structure and the habit of the worms them- 
selves, the lowest being in Cerebratulus with a rate of only 5 to 
g centimeters per second and the highest in Bispira with a rate 
of 7 meters persecond. But there seem also to be differences 
in rate between species in which little difference in structure 
and habits exists, as in the cases of Eunice and Nereis (tables 
IX, X and XI), the former showing a rate of 466 cm. per sec- 
ond, the latter a rate of 165 cm. per second. 

In the species which permitted of determining the rate of 
the impulse in both directions of the cord (Nereis, table XI; 
Lumbriconereis, table XIII, and Glycera, table XIV) there 
seems to be no constant difference between the antero-posterior 
and the postero-anterior rates. : 

A question of considerable interest is whether the rate in 
the ventral nerve cord as determined in the present work repre- 
sents the rate through continuous nerve fibers as in a verte- 
brate muscle-nerve preparation or whether the nervous paths. 
are more complex. Our knowledge of the structure of the 
ventral nerve cord of the worms has been greatly advanced 
within the last few years through the researches of RETZIUs. 
(10, 1891, 1892, 1894, 1898, 1g00), Ruope (11, 1891), Brep- 
ERMANN (12, 1891), LENHOSSEK (13, 1892), FREIDLANDER (14, 
1894), ApAruy (15, 1897) and Haver (16, 1899). But while 
many of the histological elements have thus been worked out and 
found to be markedly uniform throughout the phylum, it seems 
that the number of segments with which a motor neuron comes. 
in direct relation and the extent and relation of the small sen- 
sory fibers in the longitudinal tracts and of the central or ‘‘as- 
sociation cells’’ are yet undetermined. Even the applicability 
of the neuron conception to the nervous system of the worms 


JENKINS AND Carson, Nervous Impulse in Worms. 287 


is called in question by ApAtTuy’s work. On histological 
grounds we cannot therefore say whether the ventral nerve cord 
contains such a direct nervous path extending throughout its 
whole length. 

Nor does the present work furnish anything conclusive on 
this point. It is true that the muscle-nerve-cord preparations 
of Bispira, Glycera, Eunice, Nereis and Lumbriconereis re- 
spond to a single induced shock of low intensity ; and these re- 
sponses on stimulating the cord 20 to 30 cm. from the muscle 
in Bispira and Glycera approached the uniformity of the re- 
sponses of the frog’s gastrocnemius on stimulation of the sciatic 
nerve. And such uniformity in the response to single induced 
shocks has not been found to obtain in the vertebrates when 
complex nervous mechanisms are involved. On the other 
hand, such great variation of the rate in the same preparation as 
was observed in Lumbriconereis (page 281) can not be recon- 
ciled with the physiology of simple muscle nerve preparations 
as it is known in vertebrates and molluscs. In Cerebratulus, 
Aulastomum, Cirratulus, Arenicola, and Aphrodite the muscu- 
lar response to a single induced shock applied to the nerve cord 
subsides few centimeters in either direction from the point of 
stimulation. It requires a series of shocks to obtain the mus- 
cular response if the length of nerve cord involved comprises 
many segments; which fact suggests a complex nervous path. 
But while the response of the nerve cord toa single induced 
shock points in some species to a complex, in others to a sim- 
ple nervous mechanism, the question remains yet undecided 
whether in the latter cases the nervous part of the muscle-nerve- 
cord preparation is as simple as that of an ordinary muscle- 
nerve preparation. 

It is of some interest to know that the lowest rate in the 
Annelids, that of Aphrodite, or 54 cm. per sec., is higher than 
the rate in the pedal nerve of the slug, Ariolimax, which we 
found to be 40 cm. per sec., while the rate in Glycera (430 cm. 
per sec.), Eunice (460 cm. per sec.) and Bispira (694 cm. per 
sec.) is as high and even higher than the rate in the pallial 
nerve of the swift moving Loligo (17, 1903). 


288 


Ue 


10. 


JOURNAL OF COMPARATIVE NEUROLOGY. 


LIST OF PAPERS CITED IN, THE TEXT: 


Biedermann, W. 
Zur Physiologie der glatten Muskeln. Archiv fiir die gesamte Phystolo- 
gie, XLVI. 1889. 
First M. 


Zur Physiologie der glatten Muskeln. Archiv fiir die gesamte Phystolo- 
gie, XLVI. 1889. 


von Uexkill, J. 
Zur Muskel- und Nerven- physiologie von Sipunculus nudus. Zettschrift 
fiir Biologie, XX XIII. 1896. 
Bottazzi, Ph. 


Contributions to the Physiology of Unstriated Muscular Tissue. /oxurnal 
of Physiology, Vol. XXII. 1898. 


Straub, W. 
Zur Muskelphysiologie des Regenwurms. Archiv fiir die gesamte Ph ys- 
zologie, LXXIX. 1900. 


Buddington, R.A. 

Some Physiological Characteristics of Annelid Muscle. American Jour- 

nal of Physiology, Vol. VII. 1902. 
Loeb, J. 

Beitrage zur Gehirnphysiologie der Anneliden. Archiv fiir die gesamte 

Phystologie, XVI. 1894. 
Friedlander, B. 

Beitrage zur Physiologie des Centralnervensystems und der Bewegungs- 
mechanismus der Regenwiirmer. Archiv fiir dte gesamte Physiologie, 
LXVIII. 1894. 

Maxwell, S. 

Beitrage zur Gehirnphysiologie der Anneliden. <Archiz fiir die gesamte 

Phystologte, LXVII. 1897. 
Retzius, G. 

Zur Kentniss des centralen Nervensystems der Wiirmer. Bologische Un- 
tersuchungen, N.F. II. 1801. : 

Das Nervensystem der Lumbricinen. Szologische Untersuchungen, N. F. 
TIE Yrs92: 

Das sensiblen Nervensystem der Polychaeten. Azologesche Untersuchun- 
gen. N. F.1V: 

Zur Kentniss des sensiblen Nervensystems der Hirudineen.  Arologische 
Untersuchungen, N. F. VIL. 1898. 


Zur Kentniss des sensiblen und des sensorischen Nervensystems der 
Wiirmer und Mollusken. Prologische Untersuchungen, N. F. 1X. 1900. 


ae 


12. 


13. 


14. 


15. 


16. 


UTE 


JENKINS AND Carson, Nervous Impulse in Worms. 289 


Rohde, E. 

Histologische Untersuchungen iiber das Nervensystem der Hirudineen. 

Schneider's Zoologische Beitrége, 111. 18Ql. 
Biedermann, W. 

Ueber den Ursprung und die Endigungsweise der Nerven in den Ganglien 
der Wirbelloser Thiere. /enatsche Zertschrift fiir Natturwissenschaft 
(NEWORs) OXOVELEI eT Soins 

Lenhossék, M. 

Ursprung, Verlauf, und Endigung der sensiblen Nervenfassern bei Lum- 

bricus. Archiv fiir Mikroskoptsche Anatomie, XXXIX. 1892. 
Friedlander, B. 

Altes und Neues zur Histologie des Bauchstranges des Regenwiirms. 
Lettschrift fiir Wissenschaftliche Zoologie, LXVIII. 1894. 

Apathy, S. 


Das leitende Element des Nervensystems und seine topographischen 
Bezichungen zuden Zellen. Wittheilungen aus der zoologische Station su 
Neapel, 1897, XII. 1897. 


Havet, J. 
Structure du Systéme nerveux des Annélides, Nepheles, Hirudo, Lum- 
briculus, Lumbricus. Za Cellule, T. XVII. 1899. 


Jenkins, O. P. and Carlson, A. J. 
The Rate of the Nervous Impulse in Certain Molluscs. American Journal 
of Phystology, Vol. VII, 251-268. 1903. 


NOTES ON THE TECHNIQUE OF WEIGERT’S METH- 
OD “FOR STAINING MEDULLATED:> NERVE- 
FIBERS. 


By OLIvER S. STRONG, 


Columbia University, New York. 


The following is not the outcome of a systematic investi- 
gation of the technique of the method of WeiGeErtT, but simply 
of experiments made at various times to secure the best results 
on material hardened as described below. Notes were often 
made which have been collated and the conclusions pre- 
sented, not so much as invariable improvements on the regular 
procedure, but as suggestions for securing better results in cer- 
tain cases. Inthe estimation of the success of a WEIGERT 
preparation, especial weight has been attached to the number 
of fibers revealed in the gray and the brilliancy with which they 
are demonstrated. 

The material consisted of portions of the central nervous 
system, especially the cord, of the human foetus, infant and 
adult. This.material was fixed in formalin, in potassium bichro- 
mate and formalin, or in copper bichromate. The fixation in 
formalin was usually accomplished by injecting into the blood 
vessels formalin I vol. + water several volumes. After re- 
moval the brain and cord were kept in formalin 1 vol. + 
water g vols. until used. Some material which yielded fine 
preparations had been thus kept in formalin for three years. 
The material fixed in potassium bichromate and formalin was 
usually fixed by an injection zz sztu of potassium bichromate 
5%, or stronger, several vols. + formalin 1 vol. After removal, 
the cord and portions of brain were further hardened in potas- 
sium bichromate 5 % g vols. + formalin 1 vol. about a week and 
in potassium bichromate 5% alone for ten days to 2 weeks or 
so. Finally, some material, the brain and cord of a 7 months 


292 JOURNAL OF COMPARATIVE NEUROLOGY. 


foetus, was fixed by injection zz sztu of copper bichromate 5 % 
1 vol. + formalin 1 vol and further hardened, after removal, in 
copper bichromate 3% 9g vols. + formalin 1 vol. for about a 
week. A method pursued with some selachian material for the 
study of the cranial nerves will be described below. 

In order to give an idea of the various combinations of 
mordanting and decolorizing tried upon the above material a 
condensed list is given below together with the general charac- 
ter of the results. The staining fluid used was either WEIGERT’S 
alkaline haematoxylin (abbreviated in the list to ‘‘A. h.”’) ora 
neutral haematoxylin solution composed of 1 vol. of 10% solu- 
tion of haematoxylin in absolute alcohol + g vols. of water 
(‘‘N. h.”’). Where the kind of haematoxylin solution used is 
uncertain the abbreviation is ‘‘H.’’ The word ‘‘osmic”’ indi- 
cates that the section was placed in osmic acid. When this was 
done after the removal from the haematoxylin the section was 
simply rinsed in water, placed in 4% % + solution of osmic acid 
for 4% to I minute and then rinsed again in water, previously to 
decolorization. 


A. Celloidin sections of material (principally cord of infant and of 
eight months foetus) fixed and hardened in potassium bichro- 
mate + formalin as above described : 

Copper acetate. A.h. Pal. Failure. 

2.. Copper acetate. A. bh.; H. and N. h. Osmic. Pale iia. 

sults often good, in some cases fine preparations. 

3. Copper acetate. Osmic. A. h. Pal. Not very good. 

4. Copper acetate. Osmic. A. h. Osmic. Pal. Fine results. 

5. Copper acetate. H. Borax-ferricyanide. Not very good 

results—distinctly inferior to the Pal. 

6. Copper acetate. H. Osmic. Borax-ferricyanide. Some- 

what better than 5. 
7. Copper bichromate. A.h. and N. h. Borax-ferricyanide. 
Not good. 

8. Copper bichromate. H. Pal. Good. 

Copper bichromate. H. Osmic. Pal. Very good. This 
gave the best results of any combination. 

to. Copper bichromate. N.h. Ironalum. Failure. 

11. Copper bichromate-+ osmic. A. h. Pal. Failure. 


Lal 


12. 


ee 
14. 
Le 
16. 


Ea: 
18. 


19. 


20. 
20. 
22. 
2ae 
24. 


25 
2Ge 


ay 
Blocks of tissue, hardened like A, imbedded in celloidin, mor- 


StronG, Technique of Weigert’s Method. 293 


\ 

Copper bichromate. Potassium bichromate + osmic. H. 
Pal. Not good. 

Iron alum. A.h. Ironalum. Failure. 

Iron alum. H. Osmic. Pal. Failure. 

Iron alum. H. Borax-ferricyanide. Failure. 

Iron alum. Potassium bichromate + osmic. H. Pal. 
Failure. 

Potassium bichromate. A.h. and N.h. Pal. Not good. 

Potassium bichromate. A.h. Borax-ferricyanide. Not good. 

Potassium bichromate. Iron alum. H. Borax-ferricyanide. 
Fair. 

Potassium bichromate. Iron alum. H. Osmic. Pal. Good. 

Potassium bichromate. Iron alum. H. Pal. Not good. 

Potassium bichromate. fron alum. H. Iron alum. Failure. 

Potassium bichromate + chrome alum. H. Pal. Not good. 

Potassium bichromate ++ chrome alum. H._ Borax-ferri- 
cyanide. Failure. 

Potassium bichromate. Copper acetate. N.h. Osmic. Pal. 
Fair. 

Potassium bichromate. Copper acetate. N. h. Borax- 
ferricyanide. Failure. 

Potassium bichromate + osmic. N. h. Pal. Not good. 


danted z” fofo in copper bichromate several days. The sec- 
tions made were then treated as follows: 


ive 


2. 


3- 


A. h. (at higher temperature part of time) Osmic. Pal. 
Fine. 

A. h. (at higher temperature part of time.) Pal. Not very 
good, lacking in brilliancy. 

A. h. (at higher temperature part of time.) Borax-ferri- 
cyanide. Failure. 


C. Cord of a seven months foetus hardened in copper _ bichro- 


IEA PW N 


mate ++ formalin as above described. The celloidin sections 
made were treated as follows : 
- Calcium bichromate (5%, 13 days). A. h. Osmic. Pal. Fine. 


Copper acetate. H. Pal. Net good. 

Copper acetate. H.Osmic. Pal. Good. 

Copper acetate. H. Borax-ferricyanide. Failure. 

Copper bichromate. H. Borax-ferricyanide. Not good. 
Copper bichromate. H. Osmic. Pal. Good. 

No further mordanting. H. Osmic. Pal. Good. These 


204 


JouRNAL OF ComPaARATIVE NEUROLOGY. 


preparations were in many cases practically as good as 
those that were re-mordanted, there being only a slight 
diminution in the intensity of the stain. 


D. Cerebrum of infant. Hardened in formalin as described above. 


Imbedded in celloidin without mordanting. Sections made 
were treated as follows: 
I 


Copper acetate 46 hours. A.h.1% hours. Osmic. Pal. 
Not good, fibers very pale. 

Copper acetate 46 hours. A.h.1% hours. Borax-ferri- 
cyanide. Not good. 

Copper bichromate 46 hours. A.h.1% hours. Osmic. 
Pal. Not good. Somewhat better than 1. 

Copper bichromate 46 hours. A.h. 1% hours. Borax- 
ferricyanide. Not good. 

Copper bichromate 46 hours. A.h 1% hours. Osmic. 
Borax-ferricyanide. Not good. 

Copper bichromate 46 hours. A.h. 47 hours. Osmic. 
Pal. Better than any of preceding. 

Copper bichromate 46 hours. A.h. 23 hours. Osmic. 
Pal. Like 6. 

Copper bichromate 46 hours A.h. 23 hours.  Borax- 
ferricyanide. Not good. 
Even the best of the above were too pale, perhaps partly, 
but hardly principally, attributable to the youth of the 
brain. 


E. Cord and medulla of infant 1% months old. Hardened in form- 


alin as above described. Pieces mordanted i” foto, 
immediately after removal from formalin, in copper 
bichromate 3% for seven or eight days. Imbedded in 
STEPANOW’S Clove oil celloidin.! 
Sections were treated as follows: 

No further mordant. A.h. 12 hours. Osmic. Pal. Good. 

No further mordant. N.h. 16 hours. Osmic. Pal. Very 
good. . 

Copper bichromate 4 hours. N.h. 15 hours. Osmic. Pal. 
Very fine, better than 1 and 2. 

Copper bichromate 18 hours. N.h. and A. h. 6 hours. 
Osmic. Pal. Slightly inferior to 3. 


1 Zeitschrift fiir wiss. Mikroskopie, BA. XVII, H 2, 1900. 


StronG, Technique .of Wergert's Method. 295 


5. Copper bichromate 18 hours. N.h., kept slightly warmed, 
2to4 hours. Osmic. Pal. Very fine, possibly some- 
what better than 3. 

F. Cord, medulla and basal ganglia of child 3 years and 44% months 
old. Hardened in formalin as above described and pre - 
served in formalin for about three years before being 
used. 

(a) Pieces of cord mordanted in copper bichromate 3% seven days. 
Imbedded in clove oil celloidin. Sections treated as 
follows : 

1. No further mordanting. H. Osmic. Pal. Good. 

2. Copper bichromate. H. Osmic. Pal. Better than 1. 

3. Potassium bichromate-+ osmic. H. Osmic. Pal. Not good. 
(2) Pieces of cord mordanted in potassium bichromate + osmic. 
Sections made as in (a) and treated as follows: 

1. No further mordanting. H. Osmic. Pal. Much better 
than (a2) 3 but not nearly as good as (a2)1 or 2. Med- 
ullary sheaths of large fibers appear to be better fixed, 
however, than in (a). 

2. Potassium bichromate +osmic. H. Osmic. Pal. Better 
than (6) 1. 

3. Copper bichromate. H.Osmic. Pal. Very good, not so 
brilliant as (a) 2 but the fibers, especially in the white 
matter, look better than in (a). 

(c) Pieces of cord mordanted seven days in potassium bichro- 
mate 5 + chrome alum 2+ water 100. Sections made 
as in (a): 

1. No further mordanting. H.12 hours. Osmic. Pal. Fair. 

(a2) Pieces of cord mordanted seven days in potassium bichro- 
mate 5%. Sections made as in (a). 

1. No further mordanting. H.12 hours. Osmic. Pal. Not 
as good as (c). 

(ec) Pieces of cord mordanted seven days in copper bichromate 
3%. Sections made as in (a). 

1. No further mordanting. H.12 hours. Osmic. Pal. Fair, 
a little better than (c). 

(f) Pieces of cord mordanted seven days in potassium bichro- 
mate + osmic. Sections made as in (a). 

1. No further mordanting. H.12 hours. Osmic. Pal. Fair. 

(g) Pieces of cord mordanted seven days in iron alum 4%. 
Sections made as in (a). 


296 JOURNAL OF COMPARATIVE NEUROLOGY. 


1. No further mordanting. H. 12 hours. Osmic. Pal. Not 
good. 

(A) Pieces of cord mordanted eight days in copper bichromate. 
Sections made as in (a). 

1. No further mordanting. H.Osmic. Pal. Fair. 

2. Copper bichromate several hours (kept slightly warmed). 
H. 22 hours (first half hour in paraffine water-bath). 
Osmic: Pal. ‘Fine. 

3. Potassium bichromate-+ osmic several hours, etc., like 2. 
Not good, blurred and indistinct. 

{| (c) to (A) inclusive, taken from some incomplete notes made by 
my student, C. E. Doran]. 

(¢) Pieces of medulla and basal ganglia mordanted 7” fofo in 
copper bichromate 3% for seven or eight days. Sec- 
tions made as in (a). 

1. No further mordanting. N. h. 12 hours+Osmic. Pal. 
This invariably yielded fine results with a series extend- 
ing through the medulla and basal ganglia, practically as 
good as when remordanting in section was resorted to 
and without the added danger of brittleness. 


The following modification of the WeEIGERT method was 
devised by the writer in 1897 and has been found to be espec- 
ially adapted for his work upon the cranial nerves of Squalus 
acanthias. The head of an embryo at birth, cut longitudinally 
vertically in two pieces so as to just open one side of the cranial 
cavity, was hardened for about two weeks in iron alum 5% 
g vols. + formalin 1 vol. This fluid also decalcified. Although 
the material was somewhat brittle, a complete series of paraffine 
sections was successfully made. After fixing on the slide and 
successive removal of the paraffine and xylol, the slides were taken 
from the absolute alcohol and a thin solution of celloidin poured 
over them and off, thus covering the sections with a thin film 
of celloidin and absolutely preventing their removal from the 
slide during subsequent manipulations. This device, which had 
been used by the writer for some time previously to this, very 
much minimizes the danger of loss in staining serial paraffine 
sections. 

The sections were then stained, without further mordant- 


StronG, Technique of Weigert's Method. 297 


ing, in the above-mentioned neutral haematoxylin for 4 to 12 
hrs., the shorter period being found sufficient. As in all WeEI- 
GERT methods, old or used solutions of haematoxylin must be 
avoided. The sections are then decolorized in 1% or 2% iron 
alum, the decolorization proceeding slowly and evenly. With 
any degree of care, over-decolorization is easily avoided. After 
decolorization, the sections, now being a faint pinkish hue, are 
washed, dehydrated, cleared and mounted. 

With this method the peripheral nerves in Squalus were well 
fixed and stained deep blue. The color was completely removed 
from all other tissues except the denticles and sometimes por- 
tions of cartilage. The central nervous system was not as well 
fixed nor stained, but presented a fairly good WEIGERT picture. 

The decalcifying power, simplicity and certainty of this 
method recommend it for such work. The tendency of the 
iron alum-formalin to overharden and make the tissue difficult 
to section is perhaps the principal defect. With the loose tis- 
sues of the young shark this objection was not realized as it 
would be with other objects. 

This method has been reported by Herrick (State Hos- 
pitals Bulletin of N. Y. State, Oct., 1897, p. 27 and Journal of 
Comparative Neurology, Vol. VIII, Nos. 1-2, July, 1898.). 

Conclusions. Though many of the combinations in the 
above lists have not been sufficiently tested, yet from these data 
and from other observations of the writer, the following con- 
clusions which may in cases be of practical value may be 
drawn. 

(2) Fixation and hardening in formalin alone appears to be 
preferable in some respects to fixation and hardening in potas- 
sium bichromate + formalin followed by further hardening in 
potassium bichromate. The latter method is capable of yielding 
fine preparations, but at times it appears difficult to secure them 
by the ordinary procedures. Furthermore. in such material 
the medullary sheaths when stained often exhibit a vacuolated 
appearance, due apparently toa staining of the neurokeratin 
network, which detracts from the brilliancy of the preparations. 


.298 JOURNAL OF COMPARATIVE NEUROLOGY. 


It is a question whether the plain bichromate fixing and harden- 
ing be not superior in some regards. 

(0) Formalin fixed and hardened material will apparently 
keep indefinitely in formalin and yet be capable of giving fine 
WEIGERT preparations when subjected to the appropriate treat- 
ment (vida supra, F). Asa corollary to this it would appear 
advisable to keep all material to be used for WEIGERT prepara- 
tions, however fixed and hardened, in formalin instead of alco- 
hol. Material kept long in alcohol will not usually yield brilliant 
WEIGERT preparations. 

It may be well to call attention here to the fact that sec- 
tions already stained for WEIGERT but not decolorized may often 
be kept for months in water containing some formalin (to pre- 
vent the growth of molds etc.) and yet give good preparations 
when decolorized. This proved to be the case with some sec- 
tions prepared as indicated above under F (2). There is appar- 
ently a very slight gradual loss in brilliancy. Sections stained 
asin F (2) and kept thus 1%-yrs. in 10% + formalin still 
yielded good preparations when decolorized. Some of these 
which had been brought, after rinsing in water, into formalin 
and the formalin changed when discolored were practically in- 
distinguishable, when decolorized, from those decolorized im- 
mediately after staining. 

(c) Material fixed and hardened in formalin should in all 
cases be mordanted: zz Zofo before bringing it into alcohol pre- 
paratory to imbedding. Here lies the explanation, probably, 
of the discrepancy between the results obtained with formalin 
material by Borron' and those obtained by HErrIck.? BoLTon 
mordanted sections made from frozen or gum imbedded material, 
while HErrRIcK mordanted paraffine sections. It is also better 
to mordant pieces of the material immediately before they 
are to be used so that mordanted blocks need not be kept long 
before cutting and staining. 

It is probable, from this, that some of the apparently bad 
fixation of ‘medullated nerve fibers seen at times in formalin 
material is really due more to the subsequent treatment, the 


1 Journ, Anat. Physiol. Vol. XII, N. S., 1898. 2 Op. cit. 


STRONG, Technique of Weigert’'s Method. 299 


formalin fixed fibers not being so resistant as fibers impregnated 
with a metallic salt in the act of fixation and hardening. This 
view would seem to be further confirmed by the observations 
above noted (F (4)) where the fibers appeared to be in better 
histological condition when the formalin material was first 
treated with potassium bichromate + osmic than when treated 
in other ways. The superiority of osmic acid in faithfully fix- 
ing medullated nerve-fibers is well known. 

(2) Preparations from pieces mordanted zx foto can often 
be made more brilliant by remordanting the sections before 
staining. When the points noted in (c) have been observed, 
however, this may often be omitted, the gain being insignifi- 
cant and the sections being liable to become brittle. 

(e) The best mordant for WEIGERT-PAL preparations of 
both formalin hardened and other material is probably bichro- 
mate of copper, this apparently excelling WeEIGERT’s chrome 
alum-bichromate mixture. 

Bichromate of copper is a reagent which has some qualities 
to recommend it in neurological work, It was first used by the 
- writer in 1898 and had not been previously applied in histolog- 
ical work as faras I am aware. It is an energetic fixing and 
hardening reagent, more so apparently than any other bichro- 
mate. This indicates its use, either with or without formalin, 
in fixing and hardening foetal brains and spinal cords and as a 
mordant which gives the most brilliant results when the WEI- 
GERT-PaL method is to be used. Asa hardening fluid, the re- 
sults here are confined practically to the above-mentioned (C) 
cord of a 7 months foetus. The material and fibers were per- 
haps somewhat overhardened and shrunken but yielded very 
fine preparations. One hasty trial of this reagent in the method 
of Gore did not produce noteworthy results. 

(f/) Copper bichromate and other bichromates do not give 
good results as mordants (unless, as in the usual method, there 
is remordanting with copper acetate) where the borax-ferricy- 
anide decolorizer is to be used. 

(g) Copper acetate asa mordant invariably yields poor 


300 JOURNAL OF COMPARATIVE NEUROLOGY. 


results with the Pat decolorization unless osmic acid is used as 


indicated below (£). 
(2) Iron alum did not prove to bea valuable mordant— 


except where used asa hardening reagent for the peripheral 
nerves as above described. It was still worse as a decolorizer. 

(¢) Mordanting the celloidin block zx ¢ofo without further 
mordanting does not give such brilliant preparations as when 
the celloidin sections are mordanted instead. 

(7) No constant differences were noted between the WEI- 
GERT alkaline haematoxylin and the neutral haematoxylin. Cer- 
tainly with the Pat decolorization the neutral is as good or possi- 


bly somewhat better. 
(£) A slight or considerable increase in the brilliancy of 


WEIGERT-PAL preparations can often be obtained by dipping 
the sections in osmicacid for a fraction of a minute immediately 
after they have been removed from haematoxylin and rinsed in 
water. When this is done, sections mordanted in copper acetate 
before staining will often give good results with the Pat decolor- 
ization, though otherwise, as indicated above (g), they would be 


useless for this method. 
Osmic, used in this way before the borax-ferricyanide decol- 


orization always gave poor results. 

(7) The time of mordanting and staining is omitted in A, 
B and C for purposes of condensation. When the time devoted 
to these processes has been too short, the stain is too pale and 
the finer fibers are not be well demonstrated; when too long, de- 
colorization is too protracted and the ground will not be suffi- 
ciently decolorized or the fibers will be over-decolorized. Nat- 
urally, the longer the mordantage the shorter the time required 
for staining. Usually, at the temperature of the room, about 
12 hrs. for mordanting and about 4 to 6 hrs. for staining were the 
most favorable, but no rule can be laid down, owing to the 
condition of the material and other factors. Heat will of course 
accelerate the process but this is not necessarily to be recom- 
mended, it being liable, especially when mordanting with copper 


bichromate, to render the sections brittle. 


Dept. of Zoology, . 
Nov., 1903. 


DHE SDOCTRINE OF NERVE COMPONENTS AND 
SOME, OF ITS: APPLICATIONS:* 


By C. Jupson HERRICK. 


The original purpose of the students of nerve components 
was the analysis of the peripheral nervous system into units 
which should have at the same time a functional and a struc- 
tural significance. This obviously is not the case with the 
cranial and spinal nervesas commonly enumerated. The struc- 
tural peculiarities of each of the twelve pairs of cranial nerves, 
for instance, while fairly well defined in the human body, are 
very diverse in the vertebrate series as a whole. Thus the facial 
nerve from being predominantly sensory in lower vertebrates 
(more than half of its fibers in fishes belonging to a sensory 
system not represented at all in mammals) becomes in man pre- 
dominantly motor with only a vestigeal remnant of the sensory 
components, and even the motor component innervates chiefly 
muscles new to the. mammalia. We might multiply illustrations 
of the structural instability of the cranial nerves. And that 
the cranial nerves have any special significance as functional 
units cannot be maintained for a moment, no two pairs in the 
human body having even approximately the same function. 

But the first measurably complete analysis of the cranial 
nerves into their components for their entire extent showed at 
once the presence of certain structural and functional systems 
of components, the laws of whose distribution have apparently 
little to do with the serial order of the cranial nerves as com- 
monly enumerated. 

We have, then, a number of systems of components each 
of which is defined structurally by similarity of peripheral and 


1 Presidential Address delivered before the Ohio State Academy of Science, 
Nov. 27, 1903. 


302 JOURNAL OF COMPARATIVE NEUROLOGY. 


central terminal relations, and functionally by the trasmission of 
nervous impulses of the same type or modality. Among these 
systems are tactile, auditory, visual, olfactory, motor, gusta- 
tory, etc., each with very characteristic terminal relations. 

Now, this structure is absolutely meaningless apart from 
its function. Let any one who doubts this spend a few months. 
(as I have done) in trying to master and correlate the existing 
literature of the cranial nerves of vertebrates. Though these 
descriptions were for the most part written by famous masters 
of anatomical science, yet in their aggregate they present an 
indigestible mass of confused and meaningless detail, crude fact, 
well spiced with error, for the most part not worth the prodig- 
eous labor of digging it out of the oblivion of classic tomes of 
by-gone anatomists. 

I do not mean to imply that all the problems of cranial 
nerve morphology are now cleared up; but I do claim that 
there is no longer any necessity for the further accumulation of 
uncritical and meaningless fact in this field of research. We 
have already gone far enough to point the way toward certain) 
lines of fruitful correlation. We can not only correlate structure 
with structure, but we can interpret structure by function and. 
thus bring out a fuller meaning. We are at least coming into a 
realization of the fact that we cannot fully understand any struc- 
ture until we know what it can do. 

This point of view of course is not new, but as worked out 
practically in the peripheral nervous system it is exerting a 
clarifying influence upon our knowledge of the central system 
also. The present demand in cerebral anatomy is for conduc- 
tion paths, for functional systems of neurones, and precise 
knowledge of the pathways between the brain and the periphery 
is the first step in such a central analysis. 

The primary function of the nervous system is to facilitate 
the reaction of the organism to the external forces of the environ- 
ment. Later, as the reacting mechanism becomes more com- 
plicated, the nervous system assumes the function of coordi- 
nating this mechanism, i. e., of reaction to the forces of the 
internal environment. These two functions lie at the basis of 


HERRICK, (Verve Components. 303 


our most fundamental division in the analysis of the nervous 
system ; viz.: (1) the somatic systems (sensory and motor) for 
bodily responses to external stimuli, and (2) the visceral systems 
(sensory and motor) for visceral reactions to internal stimuli. 

Each of these great divisions has been analyzed periph. 
erally, more or less imperfectly as yet, into systems of compo- 
nents, as suggested above. Every such system of nerve fibers 
performs a separate function, conducts a single type of nervous 
impulse, either afferent, i. e., sensory, or efferent, i. e., excito- 
motor, excito-glandular, etc. The following systems are already 
distinguishable anatomically : 


I. Somatic SYSTEMS. 

1. Tactile, or general cutaneous, 

2. Acustico-lateral, including nerves for lateral line organs (in the Ich- 
thyopsida) and for organs of equilibration and hearing (in verte- 
brates generally). These organs and their nerves have probably 
been derived phylogenetically from the general cutaneous system 
and, like the organs of the latter type, are adapted for the recep- 
tion of various kinds of mechanical impact, either rhythmic or 
non-rhythmic. 

3. Visual (a system of uncertain relationships, provisionally classified 
under the somatic sensory). 

4. Somatic motor, for the innervation of skeletal or voluntary muscles. 

II. VIscERAL SYSTEMS. 

5:  Vusceral sensory, unspecialized sensory nerves of the viscera, distrib- 
uted chiefly through the sympathetic nerves. 

6. Gustatory, innervating specialized sense organs (taste buds) of chem- 
ical sense probably derived phylogenetically from the preceding 
type. 

7. Olfactory (provisionally classified here because of the apparent resem- 
blance between taste and smell). 

8. Visceral motor, distributed chiefly to unstriped and involuntary mus- 
cles, generally through the sympathetic system. 

9. ELxcito-glandular, provisionally classified here because of genera — 


resemblance to the last mentioned type. 
There are numerous other systems which can be differen- 


tiated physiologically, but which cannot as yet be completely 
separated anatomically and classified, such as nerves for the 


304 JoURNAL OF COMPARATIVE NEUROLOGY. 


thermal sensations, muscle sensations, etc., but enough has 
been cone to enable us to lay down the general plan or pattern 
of the periphera) nervous system asa whole and to define 
the main pathways by which stimuli of different modalities 
reach the brain and are reflected back to the responsive organs. 
Our anatomical knowledge of these pathways is sufficiently well 
controlled by precise physiological experimentation to enable us 
to state with confidence that each of the nine systems mentioned 
above is a real functional unit. 

The fibers composing these systems may reach the central 
nervous system through a series of many nerve roots arranged 
in a segmental way, like the general cutaneous nerves of the 
spinal cord, or they may all be represented in a single large 
nerve, like the optic and olfactory. Thus it happens that some 
nerves, like those last mentioned, are ‘‘pure’’ nerves, while 
others, like the facialis or vagus, are ‘‘mixed,’’ containing in 
some cases as many as four anatomically distinguishable com- 
ponents. 

It is a general rule that in the body the components tend 
to be distributed among a large number of nerves in a more or 
less segmental way, while in the head they tend to be concen- 
trated into a few pathways, or only one, into the brain, an adap- 
tation which presents obvious advantages for the simplification 
and unification of the secondary reflex paths from these pri- 
mary centers. 

Now, the central nervous system is, as we have already 
seen, primarily a mechanism to facilitate the reaction of the 
animal to impressions from without, in other words, to put the 
body in correspondence with the environment. Its structure is 
directly determined by the avenues of sense through which 
these stimuli come in and by the character of the responses to 
these stimuli which are necessary for the conservation of the 
organism. In view of the fact that we already possess a de- 
tailed knowledge of these peripheral nervous pathways, it is 
manifest that we have here a most favorable avenue of approach 
in an analysis of the inconceivable complexity of cerebral 
structure. 


HERRICK, Verve Components. 305 


We must know in detail the possible reflex pathways in 
the brain for all olfactory, visual, gustatory responses, etc., in 
the vertebrate type, and then on the basis of such a functional 
subdivision of the brain the problem of the mechanisms of 
higher cerebral processes may be attacked with a reasonable 
hope of success. The investigation of the internal organiza- 
tion of the brain may be pursued in several ways: 

I. The direct study of the human brain, both normal and 
pathological. On account of the enormous practical importance 
of neurology to both human psychology and pathology, re- 
search naturally turned directly to the human brain; but a more 
unfavorable starting point could not be found. 

II. It is now generally recognized that the complex hu- 
man brain can best be understood by finding first a simpler pat- 
tern such as is presented by one of the lowest vertebrates. Ac- 
cordingly the phyletic method has dominated all recent neuro- 
logical research. The brains of individual species are studied 
and monographed, particular attention being paid to the lower 
members of the vertebrate series in the hope of finding in them 
a schema or paradigm which can be followed upward through 
the comparative anatomical series and, after comparison with 
the ontogeny of higher brains, lead to a reconstruction of the 
phylogenetic history of the brain. While this method has been 
of great service, especially to such problems as can be ap- 
proached from the study of external morphology, it is im- 
mensely difficult when applied to the histological problems, and 
as a matter of fact has not as yet taken us very far. 

III. A third method, instead of taking an entire brain as 
the unit of research, concentrates attention upon a single func- 
tional system and seeks to get exhaustive comparative knowl- 
edge of it in many types. Starting with a fairly accurate and 
detailed knowledge of the functional systems at the periphery, 
we have simply to extend the lines of inquiry here blocked out 
for us. 

This gives a type of problem which is much more ap- 
proachable than the others. It is iiot so complex but more 


306 JouRNAL OF COMPARATIVE NEUROLOGY. 


intensive. Of still more importance are the facts that the ana- 
tomical data can be directly correlated by physiological experi- 
mentation, and the method is open to experimental control all 
along the line. Our degeneration methods open up _possibili- 
ties here which are incomparably more valuable than the most 
precise anatomical observation. 

And nature has performed for us a series of experiments 
which are in a sense the converse of our degeneration methods. 
The various sensori-motor systems are very unequally developed, 
some animals possessing one ina high state of elaboration, 
some another. If therefore we begin our studies on the visual 
system for instance, with animals such as most birds with very 
highly developed eyes, and then compare with animals with 
vestigeal eyes, it is evident that we have here a means of isolat- 
ing the system for scientific study which has some points of 
superiority over artificial experimental methods. Fortunately 
within the group of fishes, whose brains are all constructed on 
a plan fundamentally similar, we have the most remarkable di- 
versity in the degree of development of the several systems, so 
that this is a favorable starting point for this method, especially 
since the brain is composed almost wholly of the simpler reflex 
mechanisms without the complications which we find in mam- 
mals due to the enormous development of higher associational 
centers in the forebrain. Some fishes have huge eyes, some 
are blind; some have elaborate olfactory apparatus, some very 
slight ; some show a marvelous hypertrophy of the organs of 
taste, or touch, etc. These organs are all open to physiologi- 
cal study and so the functions can be accurately determined. 
Then, having found the cerebral pathways for each system 
where it reaches its maximum development, we can more easily 
trace out the system in other types, and thus arrive ultimately 
at a full knowledge of its evolutionary history. 

All scientific method is both analytic and synthetic. In 
the phyletic type of neurological method, these two processes 
are apt to be far separated and the observed facts may remain 
inert and relatively meaningless, because imperfectly under- 
stood, incoordinated. In our third type of method, on the other 


Herrick, Verve Components. 307 


hand, it is easier to correlate the data as we go along, the syn- 
thesis accompanies the analysis, and the possibility of experi- 
mental control should keep the student in closer touch with his 
guiding facts and discourage general speculation. 

As a concrete illustration of the practical method of apply- 
ing the doctrine of nerve components in the functional analysis 
of the nervous system, we may summarize briefly the progress 
which has been made up to date in the study of the gustatory 
system. 

In man, as is well known, the sense of taste is not very 

highly developed. The peripheral organs, or taste buds, are 
situated chiefly on the tongue, those near its base innervated 
by the glossopharyngeal nerve, and those near the tip probably 
by the chorda tympani of the facial nerve. But the gustatory 
pathway toward the brain is very imperfectly understood and 
many points are still in controversy, while the central path is 
almost wholly unknown. 
; But in certain fishes, such as the carp and cat fish, this 
system of sense organs is enormously exaggerated. Taste buds 
are found, not only in the mouth, but all over the outer skin 
- and barblets. Direct experiment shows that these fishes actually 
do taste with these superficial sense organs—unlike some peo- 
ple, their taste is not all in their mouth. 

The experiments made on the cat fish (Ameiurus) show 
that these fishes seek their food by feeling for it with the barb- 
lets and by means of them they discriminate between edible 
and non-edible substances, that they habitually use both the 
sense of touch and the sense of taste for the purpose and that 
they can be taught to discriminate between tactile and gusta- 
tory stimuli applied to the skin and will turn and snap up sav- 
ory substances and reject objects which feel like them but are 
devoid of taste. 

The exact distribution of the gustatory sense organs has 
been determined and their nerves traced back to the brain. We 
get the gustatory reaction from the skin as described above in 
fishes which possess these cutaneous sense organs, and the reac- 


308 JOURNAL OF COMPARATIVE NEUROLOGY. 


tion is not obtained from fishes which do not possess such sense 
organs and nerves. | 

All of these cutaneous sense organs are innervated from a 
single nerve, the sensory root of the facial (corresponding to 
the portio intermedia of human anatomy), which is the biggest 
nerve in the body. The center in which this nerve terminates 
in the medulla oblongata is about as big as the entire forebrain, 
instead of being barely discernable by refined histological meth- 
ods, as in the human body. And the secondary gustatory 
path, which in man is totally unknown, is the largest single 
tract in the brain, both in the cat fish and in the carp! 

The primary gustatory center in the medulla oblongata is 
bilobed, the ‘‘ facial lobe”’ receiving the gustatory fibers from 
the skin and the ‘‘ vagal lobe 
From these lobes there is both an ascending and a descending 


” 


receiving those from the mouth. 


gustatory path. The latter passes down to the point where the 
medulla oblongata merges into the spinal cord and there termi- 
nates in a special nucleus which is intimately related to the 
funicular nuclei, a center for tactile sensations. Here the tac- 
tile and gustatory stimuli are coordinated and a comimon de- 
scending bundle (tertiary path) passes back into the spinal cord 
for the body movements necessary to turn toward the food ob- 
ject. The ascending secondary gustatory path extends upward 
to a big nucleus under the cerebellum, from which tertiary path- 
ways extend forward and downward into the midbrain (chiefly 
in the inferior lobe), then backward by a descending path of 
the fourth order into the medulla oblongata to reach the motor 
nuclei of the cranial nerves. 

We have already gone far enough into our analysis of 
these secondary and tertiary gustatory paths to make it per- 
fectly safe to predict that all of the habitual gustatory reflexes 
which we have observed in these fishes can be followed anatom- 
ically through the brain for their entire extent. | And since we 
have the strongest reasons for believing that the elementary 
reflex paths are essentially similar in mammals and fishes, we 
expect to find here an important guide for further research in 
human anatomy. 


Herrick, Nerve Components. 309 


So the other sensori-motor systems may be severally inves- 
tigated, beginning the attack in each case with some species 
low down in the, vertebrate series in which this particular 
mechanism is highly developed, and then extending the re- 
search to higher and lower types. 

We may ultimately hope for a subdivision of the brain 
which shall be both structural and functional, each organ or 
pathway being given its function or meaning in the system as a 
part of the machinery of keeping the body in vital, helpful 
contact with environing forces. The great morphological 
‘thead problems,’ such as the primitive metamerism and the 
subsequent marvelous kalaidoscopic changes in structure and 
function of the component segments, these must all be read 
through the medium of such an intensive study of these factors 
upon which all differentiation has in last analysis depended. 

There is another point of view from which I have been 
somewhat interested to develop the implications of the doctrine 
of nerve components, that of scientific methology in general. 

It is said that scientific explanation consists essentially in 
such an organization of facts that they may be generalized or 
included under certain laws or uniformities which permit a fore- 
casting of future events. Now, without going into an exposi- 
tion at this time of the implied philosophy of nature, I think 
that a little reflection will show that this statement, while true 
in a certain limited sense, is very defective. 

What is the nature of this organization of facts from which 
so great benefits are expected to flow? Can it in last analysis 
be anything other than the correlation of experience? All of 
the ‘‘facts’” with which we deal have grown up in experience; 
they are in a literal sense the products of our experience. As 
men of science we have nothing to do with ‘‘things-in-them- 
’ only with phenomena, out of which we have con- 
structed by mental process certain objective things which we 
regard as real—‘‘constructs,’’ or in common parlance, objects, 
facts, data. 


selves,’ 


By these things which grew up in experience (we have in 
most cases forgotten how) we measure up and evaluate all new 


310 JouRNAL OF COMPARATIVE NEUROLOGY. 


experience. If the new sense presentation is a yellow dog with 
white feet we assimilate it at once with previous experience and 
approve it as a valid fact. If, on the other hand, it is a green 
dog with thirteen scarlet heads each with a forked tongue, we 
are apt to ask, Am I awake or asleep? or, What was I drink- 
ing last night? Such an experience may be vividly real to me, 
but if awake and sane I do not accredit it as an object of sense, 
as a fact of experience, unless I can correlate it with the body 
of fact already approved. 

But scientific laws are merely ‘‘facts’’ of wider import, 
which rest on a foundation of broader experience such that, 
when objectified, they remain not as concrete elementary exper- 
iences but as general categories including many such elements. 
The scientific generalization or law must therefore be approved 
or evaluated in a way strictly analogous with that by which we 
test sense impressions; that is, to be acceptable it must fit in 
harmoniously with the whole content of experience—‘‘it must 
explain all the facts.”’ 

In the solution of any scientific problem that method is 
most likely to lead directly to fruitful results, other things being 
equal, which favors the correlation of the data all along the line 
so that each correlation may become at once a datum for future 
research, instead of reserving the major correlations until near 
the end of the investigation. And in biological research, to 
return to our text, we must not forget for an instant that the 
organism is a functoonmmg mechanism. We cannot hope to un- 
derstand any animal or plant or organ until we have an exhaus- 
tive knowledge of how it works. The anatomical fact is dead 
and inert unless it is vivified not only by the ‘‘salt of morpho- 
logical ideas’? as it was so happily phrased years ago, but also 
by the fresh warm blood of functional explanations. 

Anatomy has given place, within the memory of even the 
younger generation of biologists, to morphology, in which the 
explanation is indissolubly linked with the fact. Nor can we 
stop here. No anatomical fact is complete until its physiolog- 
ical significance is added thereto. Like the old-time descriptive 
anatomist, the ‘‘pure’’ morphologist (or shall we dubb him 


Herrick, Nerve Components. 311 


‘‘poor morphologist’’?) has no longer any tenable standing 
ground. What I mean is that anatomical structure cannot be 
understood as the morphology of today demands that it must 
be understood without a full knowledge of the functions of the 
parts, and we must ‘know evolution of function before we can 
have true knowledge of the evolution of structure. Andasa 
matter of fact the biological public is just now coming into a 
practical realization of the truth that we must have a compara- 
tive physiology parallel with our comparative anatomy. It 
seems to us now very strange that we have had to wait a whole 
century after the birth of comparative anatomy for even the 
beginnings of a realization in practice of this elementary 
principle. 

That researches in descriptive anatomy and in pure mor- 
phology are still necessary and will continue to be called for to 
the end of the age there can be no doubt; but it is important 
that we remember that no study of structure is complete until 
the whole significance of that structure (including the evolution- 
ary history of both its form and its function) is exposed and 
the whole complex of fact and meaning not only woven to- 
gether into a single fabric, but fitted into the great pattern of 
reality as a whole in its proper place. 

Now, no one of us can do this perfectly and, as time ad- 
vances and the totality of the known becomes ever more vast 
and intricate, the difficulty grows apace. And yet this we must 
do in some measure in so far as we hope to rank as real build- 
ers inthe permanent temple of truth. If we find ourselves 
unable to see the whole edifice in its proper perspective (as in- 
deed who can ?) we can at least build harmoniously with that 
nitch in which we find ourselves. Let no man delude himself 
with the idea that he is building for himself alone, that he builds 
on no other’s foundation or that he can with safety ignore the 
labors of his coadjutors. Let no research worker hedge him- 
self about and work in isolation ; harmonious cooperation is the 
only possible way to get that breadth of view which all lack as 
individuals. 


a2 JOURNAL OF COMPARATIVE NEUROLOGY. 


In our work on the nerve components we have endeavored 
to live up to these ideals. In so far only as we succeed in ef- 
fecting wide and stable correlations from both the anatomical 
and the physiological side can we hope to be able to build a 
structure which shall endure as a secure foundation for an ulti- 
mately complete functional subdivision of the nervous system. 


COLUMELLA AURIS AND NERVUS FACIALIS: IN 
THE URODELA.’ 


By B. F. Krinespury. 


The following communication sets forth the results of a 
study made upon the relations and development of the parts in 
the otic region of the head in Necturus maculatus, and 
in comparison with that form, Desmognathus fusca and Spel- 
erpes bilineatus. 

The need fora careful study of (1) the relations of the 
facial nerve to the columella auris in the various Urodela, and 
(2) the homology of the suspensorio-opercular connections in 
the different forms of Amphibia has been emphasized by 
Gaupp.” From a comparison of the statements of WIEDER- 
SHEIM,® Hux ey,‘ ParKeEr,°® and Hasse® he was lead to conclude 


1 This may be considered as a partial preliminary communication upon the 
development of the skull of Necturus maculatus, undertaken at the suggestion 
of Professors WIEDERSHEIM and GAUPP, in the Anatomisches Institut at Frei- 
burg. I wish to acknowledge my indebtedness to them and to Professor KEIBEL 
and others, for suggestions and material. Since the completion of this manu- 
script in May 1902, more than a year has elapsed, and in sending it to the press 
now, I take the opportunity of noticing papers that have appeared in the mean- 
time—those of KINGSLEY and COGHILL. 


298, Gaupp, E Ontogenese und Phylogenese des Schalleitenden Appa- 
parates bei den Wirbeltieren. Merkel u. Bonnet, Ergebnisse d. Anat. u. Entw., 
1898, Bd. VIII, pp. 989-1149. 


3°77, WIEDERSHEIM, R. Das Kopfskelet der Urodelen. Morph. Jahré., 
Bd. I1I, pp. 352-548. 


#74, HUXLEY, TH. H. Onthe Structure of the Skull and the Heart of 
Menobranchus lateralis. Proc. Zool. Soc., 1874. 


5?77, PARKER, W. K. On the Structure and Development of the Skull in 
the Urodelous Amphibia. Pt. I. Phzlos. Trans. Roy. Soc., Vol. 167, Pt. 2. 

82a, On the Morphology of the Skull in the Amphibia Urodela. 7Z7yvans. 
Linn. Soc., Ser. 2, Vol. I1. 

82b, On the Structure and Development of the Skull in the Urodeles. 
Trans. Zool. Soc., London, Vol. XI, pp. 171-214. 


6°73, Hasse, C. Ueber den Bau des GehGérorgans von Siredon pisciformis 
und iiber die vergleichende Anatomie des Kiefersuspensorium. Anat. Stud., 
Bd. I, No. XV. 


314 JOURNAL OF COMPARATIVE NEUROLOGY. 


that there were apparently two methods of connection of the 
operculum with the suspensorium (quadratum). Thus, WIEpD- 
ERSHEIM gives as the universal condition, that the nervus facialis 
passes above the suspensorio-opercular connection ; Hux try de- 
scribed a suspensorio-stapedial (opercular) ligament wxder the 
facial nerve; Hassr, in Siredon (Amblystoma) described the 
nerve as under the columella; while the statements of PARKER 
are not always clear, though it is evident that in the different 
Urodela both relations of columella or suspensorio-opercular 
ligament and nerve were described. 

The study of the relations in the three forms above men- 
tioned, to which Proteus anguineus, Amphiuma means, and 
Amblystoma tigrinum (larva) may. be added, has shown that in 
all except Necturus, the nervus facialis passes below (ventrad 
to or cephalad of ) the suspensorio-opercular connection. In 
Necturus, the ramus jugularis facialis passes above (dorsad to) 
the ligament, the remainder of the nerve, i. e. ramus mandibu- 
laris externus and internus, and ramus palatinus being below 
(ventrad or cephalad to) this structure. Furthermore, in these 
three forms, the columella or ligament passes from the opercu- 
lum to the bone which lies partly upon the ear capsule and 
partly upon the external surface of the quadratum—and which, 
as far as I can judge from the evidence at hand, I regard asa 
squamosum ;—and not (primarily) to the cartilage of the quad- 
ratum as heretofore stated. This isa fact of considerable mor- 
phological importance. A more detailed description of the 
relations in the forms follows : 

Necturus Maculatus.. In this form Hux.ey' described the 
‘‘suspensorio-stapedial ligament’ as arising from the ‘‘middle 
of the posterior edge of the quadratum—and passing upwards 
and backwards to the stapes. The Hyomandibular branch of 
the seventh nerve passes passes above this ligament to its dis- 
tribution just as it passes above the columella auris in the Frog.”’ 
WIEDERSHEIM made no different statement of relations. This 
structure described by Hux.ey, which was presumably a sheet of 


VOps ctit-p. 192. 


Kinespury, Columella Auris and N. Facials. 2a 


fascia, is not the true suspensorio-opercular connection, which is 
correctly described by Cope,’ as passing from the operculum to. 


fig. 7. Diagram from a drawing of the left side of a model of the skull of a 
Necturus 49.5 mm. long. 


a.—Os articulare (angulare ?). 6.—Nervus bucecalis. 
c.—Columella (operculum). d.— Os dentare. 
f.—Os frontale. A.—Ceratohyale. 


7-—Ramus jugularis VII. 

L. A.-s.—Ligamentum hyo-suspensoriale. 

L. m.-h.—Ligamentum mandibulo-hyoidale. 

L. s.-e.—Ligamentum squamoso-columellare. 

m.e. VZ/,—Ramus mandibularis externus facialis. 

m. V.—Nervus maxillo-mandibularis trigemini. 

m. z2.—Ramus mandibularis internus facialis. 

o.—Nervus opticus. o. m.—Nervus oculomotorius. 
o. p.—Ramus ophthalmicus profundus trigemini. 

o. g-—Os quadratum. 


o. s.—R. ophthalmicus superficialis facialis. p.—Os parietale. 
pp.—Os palatopterygoideum. ps.—Os parasphenoideum. 
g.—Quadratum. r. c.—Ramus communicans glossopharyngei. 
s.—Os squamosum. t,—Trabeculum. 


the squamosum, who does not however, give the relation of the 
nervus facialis. DDrtNER has recently described correctly the 
relations in both Necturus and Proteus. 


1Copr, E. D. The Batrachia of North America. Bull. U. S. Nat’l. Mu- 
seum, No. 34, 18. 


316 JOURNAL OF COMPARATIVE NEUROLOGY. 


The following description of the relations in a Necturus of 
49.5 mm. length, based in part on a model of this stage (Fig. 1), 
will serve asa basis of comparison. The operculum at this 
stage is roughly oval in outline and slighly ridged along its long 
axis. At its cephalic end it is fused with the otic capsule, pro- 
jecting backward into the fenestra vestibuli. From the cephalic 
end a dense ligament passes cephalad and dorsad to the os 
squamosum at about its middle point. The bone forms a slight 
curve, the convexity looking upwards, and it lies upon the 
external semicircular canal of the otic capsule, extending down 
over the otic process of the quadrate and becoming closely 
connected with a bone lying upon the external surface of the 
quadratum, and which it partly covers. This bone’ I shall 
describe in another place. The squamoso-opercular ligament 
is attached to the under side of the squamosum where the bone 
passes from the ear capsule to cover the outer side of the pro- 
cessus oticus quadrati. At this stage the ‘‘stapedial’’ process 
of the squamosum present in the adult has just begun to de- 
velop. The ligament, in its course from the operculum to the 
squamosum, passes external (laterad) to the ramus jugularis 
facialis and the vena jugularis. The ramus jugularis passes out- 
ward and slightly backward, between the ligament and the vein 
to the dorsal edge of the former where it receives the ramus 
communicans glossopharyngei, which lies close to the ear cap- 
sule laterad to the vena jugularis. Beyond the point of the 
union with the ramus communicans, the jugular branch of the 
seventh passes outward, under the ventral edge of the squamo- 
sum to curve around the dorsal side of the otic division of the 
M. depressor mandibuli. The ramus mandibularis externus 
facialis from its ganglion which lies immediately outside the 
foramen for the facial nerve, in a depression just caudad of the 


1 This bone arises in Necturus as a separate ossification, whose lower end 
subsequeutly is fused with or becomes the ossification of the quadrate. In 
Desmognathus and Spelerpes the same bone lies farther back, projecting under 
the squamosum, and in the adult forms the process of the quadrate named for 
the purposes of this paper the subsquamosa! process. 


Kinesspury, Columella Auris and N. Factalts. Br7 


processus basilaris quadrati, "passes forward and outward under 
the quadratum to the outer surface of the squamosum, passing 
in front of (ventrad and cephalad to) the ligament. 

The ramus palatinus which passes forward through a fora- 
men distinct from that for the rest of the facial nerve, and the 
ramus mandibularis internus which passes immediately ventrad 
from the cephalic edge of the accessory lateral line ganglion, 
do not come into close relation to the columella, but are, of 
course, morphologically below and in front of it. 

In an older Necturus, 9.4 c.m. long, the relations are as in 
the specimen just described, save that the processus ‘‘staped- 
ialis’”’ of the squamosum has attained an appreciable length, and 
the operculum possesses a short ossified stalk to which the liga- 


Fig. 2a. Section of the Necturus embryo 19 mm. in length. c.—anlage of the 
squamoso-columellar ligament; 4.—hyoid: v. 7.—vena jugularis ; 2. 7.— 
nervus jugularis ; @. #.—M. depressor mandibuli; s.—squamosum. 

Fig. 26. Same, three sections farther forward. 


ment attaches. Neither ossification appears to be an ossifica- 
tion of the ligament, but ossifications of the squamosum and 
operculum at each end of the ligament, acomplishing in that 
way the increase in length due to growth. In the adult, the 
operculum possesses an ossified process of some length joined 


318 JOURNAL OF COMPARATIVE NEUROLOGY. 


by ligament to the relatively long stapedial process of the 
squamosum. 

That the relation of ligament to squamosum is a primary 
condition in this form and not a secondary modification, is seen 
in tracing the development of these structures. In an embryo 
Ig mm. in length (Fig. 2), the ossification of the squamosum is 
just beginning as a formation ina group of cells located upon 
the external semicircular canal of the ear. It extends down 
over the otic process of the quadratum covering with its lower 
(cephalic) end the upper end of a bone which is developed upon 
the external surface of the quadrate. At this stage, the oper- 
culum is just beginning to chondrify as a distinct center, and 
from it a cord of cells is continued forward, ventral to the vena 
jugularis and the ramus jugularis, to the cell surrounding the 
developing squamosum, becoming continuous with them a short 
distance (50y) back of the processus oticus quadrati. The cells 
are of course continuous with those of the squamosum and also 
with the cells between that bone and the quadratum, so that the 
squamosum, the quadratum, and the ligament-anlage, may be 
said to be joined together by a common mass of cells. In the 
just hatched larva, likewise, the ligament-anlage, clearly goes 
to the under side of the squamosum and inserts itself between 
that bone and the processus oticus quadrati, so that it might be 
interpreted as going to both structures. As soon as the con- 
nective tissue fibers develop, however, the relation is seen to be 
with the squamosum and not with the quadratum. It is inter- 
esting to note the relatively early development of the ligament 
—practically at the same time as the squamosum and the oper- 
culum—later, however, than the chondrification of the chon- 
drocranium. 

Spelerpes bilineatus. In this form, as well as in Desmog- 
nathus, the suspensorio-opercular connection possesses the same 
relation to the nervus facialis—that is, the nerve lies entirely 
cephalad and ventrad to the stilus columellae; in other words, 
under it. In relation to the jugular vein, the stilus possesses 
the same relation as the ligament described in Necturus—i. e. 
it passes ventrad to it. 


ee 


Kincssury, Columella Auris and N. Facials. 319 


In the adult Spelerpes (Figure 3), the stilus is cartilagin- 
ous with a perichondral ossification continuous with the ossifi- 
cation of the operculum;—the cartilaginous core of the stilus, 
however, is distinct from the ring of cartilage within the 
operculum. 


i ——_——-— 


_>— 


Fig. 3. Spelerpes bilineatus, adult 67 mm. long. Section through the right 
otic capsule. c—Stilus columellae; o0—oral cavity; v. 7.—vena jugu- 
laris; @. m.—M. depressor mandibuli; S.—squamosum. 


The stilus passes forward, upward and slightly outward to 
the lower edge of the squamosum with which its cephalic end 
is joined by connective tissue (Fig. 3), and also with a small 
cartilage which lies upon the ventral edge of the squamosum. 
This cartilage extends forward for about 150 microns and is 
cylindrical. It is free at its caudal end, which articulates with 
the stilus, and fused with the ventral edge of the squamosum. 
The stilus and operculum are at about the same level. The 
former lies at first upon the dorsal side of the external semi- 
circular canal, gradually moving down to the lateral surface of 
the otic capsule, as it is traced forward. As it continues to 


320 JOURNAL OF COMPARATIVE NEUROLOGY. 


shift its position ventrally to pass to the outer surface of the 
quadratum, it becomes farther separated from the ear capsule 
leaving a space in which the quadratum appears. The ventral 
(lateral) edge of the squamosum is thin where the bone rests 
upon the ear capsule, but becomes thicker as the bone leaves 
that structure, i. e. where the stilus articulates with it, becom- 
ing thinner again as the bone applies itself to the quadratum. 


Fig. 4. Larval Spelerpes bilineatus, 43 mm. long; c.—stilus columellae ; 
v, j-—vena jugularis; #. 7.—nervus jugularis; d@. m.—M. depressor 
mandibuli; s.-~Squamosum. 

In the interval between the squamosum and the ear capsule, 

two processes of the quadratum extend backward, (1) a bony 

process applied immediately to the inner surface of the squa- 
mosum, extending back to the level of the cephalic end of the 

cartilage upon the ventral edge of the squamosum, and (2) a 

short cartilaginous process lying between the bony process and 

the ear capsule. The latter seems to be a part of the (morpho- 
logically) basilar process of the quadratum and is very short. 


Kincssury, Columella Aurts and N. Facials. 321 


Neither one comes into relation to the columella as do the cor- 
responding processes in Desmognathus. 

Larval Spelerpes of 25 mm., 35 mm., 43 mm. (Fig. 4) 
and 60 mm. in length, were examined in this connection and 
showed that the relation between columella and squamosum in 
this form (Fig. 4) is a primary one, as in Necturus. In the 25 
mm. larva, the suspensorio-opercular connection is represented 
by a cord of cells which passes from the operculum forward 
and upward to the ventral edge of the squamosum. This cell 
cord lies ventrad to the vena jugularis around which it curves, 
closely applied to the vein, compressed between it and the R. 
jugularis facialis, the relation of nerve and suspensorio-oper- 
cular connection being thus the opposite of that in Necturus. 
Compare Figs. 2 and 4. In a 35 mm. larva cartilage has 
appeared in the cord of cells, otherwise the relations are essen- 
tially the same as in the younger larva, while in the 43 mm. 
specimen ossification of the stilus has begun, continuous with 
the perichondral ossification of the operculum. 

The facial nerve, as has been said, lies entirely cephalad 
and ventrad to the suspensorio-opercular connection. The 
only branch which comes into contact with the stilus is the 
ramus jugularis which in the larva passes close to the ventral 
border of that structure. The ramus communicans glos- 
sopharyngei likewise, passes below the stilus, curving around it 
from its dorsal side in a course forward to join the facial. In 
the adult neither nerve is in as close relation to the stilus as in 
the larva. 

The origin and significance of the small cartilage applied 
to the ventral border of the squamosum is obscure because of 
the absence of transforming and young adult material. In the 
larva it is not present. 

With the exception of the R. jugularis and R. communi- 
cans, then, the suspensorio-opercular connection in Spelerpes 
has the sime morphological relations as the ligament in 
Necturus. 

Desmognathus fusca. (76 mm.) In this form it would 
seem as if, as compared with Spelerpes, the suspensorium were 


322 JOURNAL OF COMPARATIVE NEUROLOGY. 


displaced backward in relation to the operculum, so that the 
stilus is shorter, passes more directly outward and upward, and 
is joined more closely with the subsquamosal process of the 
quadrate (Fig. 5) than with the squamosum itself. It is, 


Fig. 5. Desmognathus fusca, adult, 76 mm. c—stilus columellae; ¢. /.— 
canalis lateralis; 4.—hyoid; wv. j.—vena jugularis; 7. c.—ramus com- 
municans; @. m.—depressor mandibuli; /. s,—subsquamosal process of 


quadratum ; s.— squamosum. 


however, joined to both bones by connective tissue, and with 
the cartilaginous process of the quadrate. This process is 
longer than the corresponding process in Spelerpes and is sepa- 
rated from the stilus by an interval of but (ca.) 50 w (Fig. 6). 
The squamosum and the subsquamosal process of the quadrate 
are essentially the same as in Spelerpes. Stilus and operculum 
are as in Spelerpes, though the cartilage in the columella is 
small. 


Kincssury, Columella Aurts and N. Factalis. 323 


Turning to the larval form for an interpretation of the con- 
dition in the adult, we find in a specimen 21 mm. in length, 
that the suspensorio-opercular connection is at this stage cellu- 
lar and extends from the cephalic border of the operculum to 
the squamosum as a dense cord of cells. It has the same rela- 


Fig. 6.4'Same,’three sections (75 yz) farther forward; g.—cartilaginous (colum- 
ellar) process of the quadrate. 


tion to the jugular nerve and vein as in Spelerpes, though it 
does not come into as close contact with either as in that form. 
Its cephalic end is rather difficult to determine (Fig. 7), since 
the anlage is continued forward to join the subsquamosal pro- 
cess (of the quadrate) which at this stage is a distinct bone, so 
that it may be said to be connected with both bones. There 
is, however, no direct connection with the (cartilaginous) 
quadrate, and from} the conditions in Necturus and Spelerpes, 
we are warranted, I think, in emphasizing the connection with 
the squamosum rather than that with the subsquamosal process 
of the quadrate which, in fact, is not as direct. Ina larva 33 


324 JOURNAL OF COMPARATIVE NEUROLOGY. 


mm. in length (Fig. 7), apparently approaching the period of 
transformation, the relations are as in the younger specimen 
save that cartilage has appeared in the suspensorio-opercular 
connection as a center distinct from the cartilage of the operc- 


Fig. 7a. Larval Desmognathus fusca, 33 mm. ; c.—stilus columellae; v. 7-— 
vena jugularis; #. 7.—nervus jugularis (R. communicans) ; @. m.—M. 
depressor mandibuli; #. s.—subsquamosal process of the quadrate ; 
§.—Squamosum. 


Fig. 76. Same; three sections farther forward. 
ulum. In a small adult (27 mm.), presumably but recently 


transformed, the cartilaginous stilus is connected more directly 
with the squamosum, but also by dense connective tissue with 


Kinessury, Columella Auris and N. Facialts. 325 


the subsquamosal and the short cartilaginous processes of the 
quadrate. The shifting of the attachment takes place in the 
growth of the adult rather than at the transformation of the 
larva. 

Ampliuma means (51 mm). Through the courtesy of my 
co-worker, Professor H. W. Norris, I am enabled to give here 
the following brief statement of the relations occurring in 
Amphiuma as found by him and verified by myself in his prep- 
arations. This form is interesting because it possesses a con- 
tinuous cartilaginous connection between the quadrate and oper- 
culum, as described by WrepeERsHeEIM,' Hay,” and WINsLow.’* 
This has been spoken of as the columella, and as the stapedial 
process of the quadrate. It evidently, however, represents 
both the columella (stilus columellae) and the primarily cartilag- 
inous process of the quadrate found in Desmognathus. The 
articulation in the specimen upon which this statement of rela- 
tions is based is much closer than it is in Desmognathus, and in 
older specimens undoubtedly, as described, there occurs a 
fusion of the two structures to form one continuous rod _be- 
tween the operculum and the quadratum. In this specimen, 
the stilus is a cartilaginous process of the operculum which is 
itself cartilaginous. The stilus columellae goes forward and 
slightly upward to become applied to the thickened ventral 
border of the squamosum to which it is joined by connective 
tissue. It is succeeded by the cartilaginous columellar process 
of the quadrate to which it is very closely connected. This 
process lies also against the ventral edge of the squaamosum and 
slightly on its inner side. The connection of the stilus, there- 
fore, is with the squamosum and the cartilaginous process of 
the quadrate and not at all with the ossification which (from the 
condition in the adult Desmognathus and Spelerpes) I have 
spoken of as the subsquamosal process of the os quadratum. 


NOp-cit-,, p. 502. 


2°90, Hay, O. P. The Skeletal Anatomy of Amphiuma during its earlier 
Stages. Journ. Morph., Vol. IV. 

5’98, WinsLow, G. M. The Chondrocranium in the Ichthyopsida. 7 fts 
College Studies, No. 5, 1898. 


326 JOURNAL OF COMPARATIVE NEUROLOGY. 


All branches of the facial nerve pass below the stilus 
columellae (and stapedial process of the quadrate) as has 
already been stated by Hay, instead of over it.’ 

Other Urodela. In Menopoma (Cryptobranchus) alone is 
the relation of columella to the squamosum described by 
WIEDERSHEIM,” and also by PARKER.” 

In Amblystoma I can only state that there is present in the 
larva a cord of cells, passing from the operculum to the ventral 
border of the squamosum, which from the position and rela- 
tion (dorsal) to the facial nerve is undoubtedly the anlage of 
the suspensorio-opercular connection. This relation of the 
‘‘columella’’ to the facial nerve, has already been affirmed by 
Hasse and PARKER. 

Proteus anguineus. Opportunity for studying the relations 
in this form was afforded me by the generosity of Professor 
WievERSHEIM. As might be expected from the published fig- 
ures (WIEDERSHEIM ; op. cit., Fig. 1g), the relations in Nec- 
turus and Proteus are the same. There is a strong squamoso- 
opercular ligament passing from the stapedial process of the 
squamosum to the short stilus columellae, and to this the 
branches of the facial nerve have the same relation as in Nec- 
turus; R. jugularis passes above the ligament, R. mandibularis 
externus below it. 


Nervus factalis.* 


Since the homology of the chorda tympani is closely con- 
nected with that of the relations and connections of the colum- 
ella auris, the following brief account of the course of the 
branches of the facial nerve is offered. The relations of the 
nerve in the larvae only of Desmognathus and Spelerpes have 


1 This is also in accord with KINGSLEY’s description. (7«/t's College 
Studies, No. 7, p. 3705.) 

2\Op; Cit., p: 502. 

Ops Gite bt. Lu spewed. 


* The following names of the branches of the ‘facial nerve are used: R. 
palatinus; R. jugularis (FISCHER); R. mandibularis internus (R. Alveolaris, 
FISCHER); R. mandibularis externus (R. mentalis, FISCHER). 


Kinesspury, Columella Auris and N. Factalis. 327 


been studied, as the changes at transformation introduce com- 
plexities unimportant in this connection. 

Necturus (9.4 cm.). The ganglion geniculi is intra-cranial, 
in the beginning of what might be described as a short facial 
canal, adjoining and cephalad of the cephalic division of the 
auditory nerve. From this ganglion the ramus palatinus arises 
as a small nerve which passes cephalad and ventrad through a 
separate foramen, and goes cephalad at the side of the trabe- 
cula, finally passing ventrad between the parasphenoid and the 
pterygo-palatinum to the roof of the oral cavity. The remain- 
der of the nerve passes laterad through its foramen and develops 
a second ganglion, which undoubtedly belongs to the R. man- 
dibularis externus, a part of the lateral line component. At 
this ganglion the nerve divides into two branches, R. mandibu- 
laris facialis, and R. jugularis; the former divides, as soon as it 
leaves the ganglion into the Rami mandibularis externus and 
internus. The R. jugularis passes upon the caudal side of the 
ganglion and has but little if any connection with it. Its course 
is nearly directly laterad for a short distance, passing dorsad to 
the ligament between that structure and the jugular vein ; be- 
yond the ligament, under the ventral edge of the squamosum 
it turns ventrad and caudad around the dorsal border of the otic 
division of the M. depressor mandibuli to pass under the fascia 
coveriny the lateral surface of that muscle. At the lateral border 
of the M. mylohyoideus posterior, it passes to the ventral side of 
that muscle. It innervates the M. depressor mandibuli, ceratohy- 
oideus, and mylohyoideus posterior. The Ramus communicans 
glossopharyngei passes forward from the ganglion complex of the 
IX and X and joins the R. jugularis just beyond the point where 
it emerges above the columella. The M. depressor mandibuli 
gains some at least of its innervation from fibers of the R. jug- 
ularis which pass back along the R. communicans. R. jugu- 
laris seems to be purely a motor nerve, though it is possible 
that it may have a small lateral line component. 

The R. mandibularis externus goes cephalad, laterad and 
ventrad under the ventral border of the squamosum below (in 
front of) the point of attachment of the ligament curving 


328 JOURNAL OF COMPARATIVE NEUROLOGY. 


around to the outer surface of the squamosum. After giv- 
ing a branch to the skin whose destination was undoubtedly 
the lateral line sense organs, it divides into two branches, one! 
passing farther caudad and mesad, so as to lie on the mesal side 
of the lower jaw, between the M. submaxillaris and the skin; 
the other passing to the outer side of the lower jaw. From 
these two branches, evidently the lines of sense organs called 
by me* gular, and oral (incl. angular) respectively, receive their 
innervation. It is possible that the gular division contains com- 
munis fibers as well as those destined for the lateral line organs. 
The M. submaxillaris I find to be innervated by the trigeminus 
(R. mandibularis internus V), in this supporting Miss Pratr* 
as against RuGe.' Both divisions are subcutaneous,—i. e. ex- 
ternal to all skeletal and muscular structure. 

The ramus mandibularis internus, separates from the R. 
mandibularis externus as it leaves its ganglion, and passes ven- 
trad and cephalad, on the inner (ventral) side of the quadrate 
soon passing through the suspensorio-hyoid ligament. This is 
the condition ina specimen g.4 centimeters in length. In 
younger specimens the nerve seems to lie on the outer side of 
the ligament, though very closely applied to it. Beyond the 


' This is evidently the branch described by VON PLESSEN and RABINOVICZ 
(Die Kopfnerven von Salamandra maculosa im vorgeriickten Embryostadium, 
1891) as ‘‘Begleiter des R. hyoideo-mandibularis (h. m’)’’—Hyomandibularis 
accessorius. By some these branches have been incorrectly called Rami man- 
divularis internus (alveolaris) and externus. The homology of either of these 
nerves with the chorda tympani, suggested by HERRICK in his ’94 paper (Am- 
blystoma punctatum) and accepted by KINGSLEY ’o2, for Amphiuma, can, of 
course, hardly hold now. COGHILL calls these, Rami mentales externus and 
internus. 


2 "95, KinGspury, B. F. The Lateral Line System of Sense-organs in some 
American Salamanders, and Comparison with the Dipnoans. Proc. Americ, 
Micr. Soc., Vol. XVII, 1895. 


: 98, Pratt, JULIA B. The Development of the Cartilaginous Skull and 
of the Branchial and Hypoglossal-Musculature in Necturus. A/orph. Jahrb., 
Bd. XXV, 1898. 


*°96, Rucr, G. Ueber das peripherische Gebiet des Nervus facialis bei 
Wirbelthieren. Festschrift fiir Carl Cegenbaur, 1896, pp. 195-348. 


Kincspury, Columella Auris and N. Facialis. 329 


ligament, the nerve is on the inner side of the M. depressor 
mandibuli, MECKEL’s cartilage and the os articulare successively. 
It is separated by connective tissue from the mucous membrane 
of the mouth which it gradually approaches, lying on the 
dorsal (mandibular) side of the depression’ between the hyoid 
and mandibular arches. At about the level of the caudal bor- 
der of the eye, it divides into two branches, one of which con- 
tinues forward on the inner side of the jaw, the other moves 
farther ventrad and mesad ; both, however, become compressed 
between the M. submaxillaris and the oral mucous membrane 
of the floor on the mouth between the hyoid (tongue) and the 
mandible. 

No communication occurs between this nerve and the Ra- 
mus mandibularis internus of the fifth. 

In the larvae of Spelerpes” and Desmognathus the relations 
of the four main branches of the seventh nerve are in general 
essentially as in Necturus. The Ramus jugularis, however, in- 
stead of curving around the dorsal border of the otic division 
of the depressor mandibuli as in Necturus, in Spelerpes passes 
through that division of the muscle, while in Desmognathus, it 
passes under the entzve muscle. In both Desmognathus and 
Spelerpes it contains a cutaneous— undoubtedly lateral line— 
component which was not found in Necturus. Asin Necturus the 
M. depressor mandibuli receives its innervation from fibers that 
accompany the Ramus communicans. The relation of both 
the R. jugularis and the R. communicans to the stilus colum- 
ellae has been spoken of in connection with that structure. 


l“The R. alveolaris VIL, composed wholly of communis fibers, follows 
the posterior border of the suspensorium to the angle of the jaw. Along this 
part of its course, the R. alveolaris lies mesially of the hyo-suspensorial liga- 
ment, and anteriorly of the deep pharyngeal evagination which represents the 
embryonic spiracular cleft.’’ ’02, CoGHILL, G. E. The Cranial Nerves of 
Amblystoma tigrinum. /Journ. Comp. Neurol., Vol. XII, p. 228. 


2 The branches and distribution of the facialis in the larval Spelerpes have 
been correctly given by Miss M. A, Bowers: The Peripheral Distribution of 
the Cranial Nerves of Spelerpes bilineatus. Proc. Am. Acad. Arts and Sct., 
Vol. XXXVI, 1900. 


330 JOURNAL OF COMPARATIVE NEUROLOGY. 


The Ramus mandibularis externus passes cephalad and 
laterad around the lower edge of the squamosum to its outer 
surface, where it divides into branches, as in Necturus, one of 
‘which curves ventrally over the outer,surface of the M. depres- 
sor mandibuli and its tendon to run forward upon the ventral 
surface of the M. submaxillaris. The other division runs ceph- 
alad upon the outer side of the lower jaw. Both seem to be 
purely lateral line nerves. 

The R. mandibularis internus separates from the externus 
at the cephalic border of the ganglion and goes laterad cephalad 
and ventrad immediately to the mucous membrane of the oral 
cavity between the hyoid arch and the quadrate and (farther 
cephalad) the mandible. In the first part of its course it lies 
in the connective tissue between the oral mucous membrane, 
the quadrate and the M. depressor mandibuli, the quadrate 
lying dorsally and the muscle laterally. Farther cephalad it 
passes on the inner side of the suspensorio-hyoid ligament, 
MECKEL’s cartilage and the os articulare on whose mesal side it 
divides, one branch passing through a canal in that bone to 
join the R. circumflexus V,' which at nearly the same level 
passes between the os dentare and MeEckeEL’s cartilage. This 
soon divides on emerging from its canal into the R. submax- 
illaris and R. mandibularis internus V. The remainder of the 
R. mandibularis internus VII runs forward between the mucous. 
membrane and the mandible. At the level of the appearance 
of the M. submaxillaris, it is compressed between that muscle 
and the mucous membrane of the floor of the mouth. The 
portion of the R. mandibularis internus VII which joined the 
trigeminus I was unable to trace. I was unable to trace the 
fibers of the R. mandibularis internus in any of the forms even 
into the neighborhood of taste buds. It is clear, that the R. 
mandibularis internus (alveolaris) in Urodeles has practically 
the same course, the only marked differences being that in 
Necturus, and Proteus, it does not pass through a canal in the 


1T use the name applied to the comparable nerve in the frog, believing 
them homologous. Compare, however, COGHILL, op. cit., pp. 265 and 266. 


Kinespury, Columella Auris and N. Factalis. 331 


Os articulare (angulare ?), while in Amphiuma,' (’02, KinesLey), 
Desmognathus and Spelerpes, Amblystoma, Salamandra and 
Triton, (CoGHILL, op. cit., p. 269), it occupies such a canal. 
In Necturus, Proteus, and Amphiuma (KINGSLEY) it does not 
anastomose with the Vth, while in the other forms it does. 

From the above relations it is seen that the only nerve 
which can be considered as a homologue of the chorda tympani 
is the Ramus mandibularis internus VII which goes to the mu- 
cous membrane of the floor of the mouth between the hyoid 
and mandibular arches.” This, of course, is the homology 
already advanced by Gaurp,* Srrone! and others, Gaupp from 
morphological relations, SrronG from the character of the fibers 
and their destination. AtLtis,° HERRICK® and GREEN’ have since 
seen reason todoubt the homology on the grounds of the pre- 
trematic position which the homologue of the chorda tympani 
must have, the nerve identified by them as R. mandibularis 
internus facialis being a post-spiracular nerve, and a Ramus facialis 
pretrematicus being chosen by them as the homologue of the 
chorda tympani. 


192, KINGS).EY, J. S. The Cranial Nerves of Amphiuma. T7/ts College 
Studies, No. 7, pp. 293-321. 


2 RUGE (op. cit., p. 294) recognizes this nerve as the chorda tympani though 
he does not identify it as the internal mandibular (alveolaris, FISHER) but seems 
to find that also present as a cutaneous nerve. COGHILL (op. cit.) regards it 
as a homologue of the chorda tympani. 


393, Gaupp, E. Beitrage zur Morphologie des Schidels. I. Primordial 
Cranium und Kieferbogen von Rana fusca. Morph. Arbetten herausg. VON G. 
SCHWALBE, Bad. II. 


4°95, STRONG, O. S. The Cranial Nerves of Amphibia. A Contribution 
to the Morphology of the Vertebrate Nervous System. Journ. of Morph., Vol. X- 


5’97, ALLIs, E. P. The Cranial Muscles and Cranial and first Spinal 
Nerves in Amia calva. /ourn. of Morph., Vol. XII, No. 3, 1897. 


6 99, HERRICK, C. J. The Cranial and First Spinal Nerves of Menidia ; 
a Contribution upon the Nerve Components of the Bony Fishes. /Journ. Comp. 
Neurol., Vol. 1X, 3-4. 


7‘o0, GREEN, H. A. On the Homologies of the Chorda Tympani in 
Selachians. /Journ. Comp. Neurol., Vol. X, 4- 


332 JOURNAL OF COMPARATIVE NEUROLOGY. 


The question seems to me to involve the correctness of the 
interpretation of the chorda tympani as pretrematic, and the 
homology of the mandibularis internus VII, in Menidia, Amia 
and Selachia, which appears to have a course somewhat different 
from that of the branch in Urodeles. For a comparison of the 
relations in fishes and Amphibia, the effect of the morpholog- 
ical differences in the suspension of the jaw and the value of 
the relation of nerves to skeletal structures in determining their 
homology, are involved; and for the larger question of the 
chorda tympani, the homology of the sound-transmitting appa- 
ratus in the different classes, as well; so that it seems to mea 
close consideration of homologies is yet premature. 

The pre- or post-trematic origin of the R. mandibularis 
internus in Urodeles cannot, of course, be determined, since 
the first gill cleft does not come to development. From its 
point of origin and course, it certainly could be pretrematic, as 
Coxe’ has pointed out, and it seems to me the possibility that 
this nerve represents a pre-trematic nerve such as GREEN, 
(e. g.) described in Selachia,? is worth considering. In this 
connection the different relations of the facial nerve and col- 
umella auris in Anura and Urodela must also be considered. 
There is here presented in allied forms, a difference of relation 


1°96, CoLE, F. J. On the Cranial Nerves of Chimaera monstrosa (Linn) 
with a Discussion of the Lateral Line System and of the Morphology of the 
Chorda Tympani. 7Zyvamns. Roy Soc., Edinburgh, Vol. XX XVIII, Pt. II, 
(No, 19). 


21 have already referred to the statement by COGHILL (p. 228) that this 
nerve could be considered pre-trematic. In the forms studied by me, however, 
the conditions, I believe, hardly warrant a definite conclusion. COGHILL, 
even, would regard the R. mandibularis internus in Urodela and Anura, as not 
homologous (p. 265), and this, too, seems to me rather extreme. The entire 
hyomandibular nerve in the frog crosses over and behind the columella auris 
and in Urodela under and in front of it. As stated in a previous paper (’95, 
The Structure and Morphology of the Oblongata in Fishes; Journ. Comb. 
Neurol., Vol. VII, p. 30) where I quote also the opinion of Miss PLATT to that 
effect, I feel that the origin and distribution of a nerve are of more importance 
than its course, which may vary, and consequently should not be too closely 
made the basis of homologies. We also see that the relation of a nerve to a 
muscle cannot be relied upon as a test. 


Kincssury, Columella Auris and N. Facials. 333 


of nerve to skeletal structure of extreme type. As is well 
known,' in the frog the hyomandibular nerve crosses above the 
columella and passes down behind it to its destination, whereas 
in Urodeles it passes in front of or below the same structure. 
Other cases of similar differences of relation in this region, 
mentioned in this paper, are (a), the relation of the jugular 
nerve in Necturus on the one hand and in the other salamanders 
investigated on the other hand; in the first case it passes above 
the columella (stilus columellae), in the second, below.  (b), 
Necturus also offers a difference in the relation of the internal 
mandibular branch to the quadrato-hyoid ligament. In Des- 
mognathus and Spelerpes the nerve passes on the inner side of 
the ligament; in Necturus, through the ligament, or on its 
outer side in younger individuals. Further (c), the R. jugu- 
laris in Necturus passes over the depressor mandibuli; in 
Spelerpes larvae, through it; in Desmognathus larvae, under it. 

The differences, in the last two cases at least, it seems to 
me, might possibly be explained on a more or less mechanical 
basis. The nerves (and muscles) are already developed and 
their course and positions established before the anlage of the 
columella or that of the quadrato-hyoid ligament has appeared, 
and the relations the latter structures assume when they do 
develop, has been determined for them by the position of the 
structures earlier developed. This explanation would not, of 
course, be an ultimate one. 

The nomenclature employed in the above descriptions is 
that suggested by Professor Gaupp. Columella, including 
operculum and its process, sti/us columellae, which may be 
joined to the suspensorium by an appreciable ligament—/ga- 
mentum suspensorto-columellare (operculare). \regard the sus- 
pensorio-columellar (opercular) connection in the forms studied 
as homologous. The term stilus columellae is used in describ- 


193, Gaupp, E. Beitrage zur Morphologie des Schadels. I. Primordial 
Cranium und Kieferbogen von Rana fusca. Morph. Arédeten, herausgegeben von 
G. Schwalbe, Bd. II. 

’99. ECKER’s u. R. WIEDERSHEIM’S Anatomie des Frosches, auf Grund 
eigener Untersuchungen durchaus neu bearbeitet. [II Abth. 1899, 


334 JOURNAL OF COMPARATIVE NEUROLOGY. 


ing the relations in Desmognathus and Spelerpes, in view of 
the structure in the adult, despite the fact that the ‘‘stilus”’ 
probably begins as a chondrification in the cord of cells extend- 
ing from the operculum to the squamosum. This point of a 
separate chondrification, however, has not been firmly estab- 
lished. In that case the ligamentum squamoso-columellare 
(operculare) and the stilus columellae of Spelerpes I should 
regard as homologous—despite the different relations to the 
facial nerve. 

In conclusion, I may say that the points which I wish to 
emphasize are: 

(1) The primary connection of the columella with the 
bone which I regard as the squamosum. 

(2) The different relations of the facial nerve to the 
(‘‘squamoso-opercular’’ connection) stilus columellaris in the 
frog, Necturus (and Proteus) and other Urodela. 

(3) The secondary nature of the connection of the col- 
umella with the quadrate cartilage, where such connection 
occurs. 

(4) The different relations of the ramus jugularis VII, to 
the musculus depressor mandibuli in Necturus, Spelerpes, and 
Desmognathus. 

(5) The course and relations of the R. mandibularis 
internus VII, in view of the possible homology with the™ 
chorda tympani. 

(6) The question of the value of. the relation of a nerve 
to skeletal parts and muscles, as a criterion of homology. 


Anatomisches Institute, Fretburg 1. B., May 1, 1902. 
Cornell University, Sept., 1903. 


EDITORIAL ANNOUNCEMENT. 


THE JOURNAL OF COMPARATIVE NEUROLOGY, as originally 
announced, was open to contributions in the field of compara- 
tive neurology, physiology and psychology. The founder, feel- 
ing that the time was ripe for a more thorough correlation of 
the facts in these different fields, planned to devote the Journal 
as much to the functional as to the structural study of the ner- 
vous system. During the thirteen years of the existence 
of the Journal of Comparative Neurology, the functional 
side of neurological work, although not wholly neglected, 
has received far less attention than was originally contemplated, 
chiefly on account of the continued ill-health of the Editor-in- 


' Chief, who had intended to devote himself primarily to com- 


parative psychology. 

Now, however, we are able to announce an enlargement of 
the editorial staff which will insure a more satisfactory repre- 
sentation of the functional as well as the structural aspects of 
neurology. 

The Journal will hereafter be known as the ‘‘Journal of 
Comparative Neurology and Psychology.” The organization 
of the editorial staff remains in general as before save that Dr. 
Rosert M. YERKEs of the Department of Psychology, Harvard 
University, will be the responsible editor for the department of 
Animal Behavior. He will be supported on the editorial board 
by representative students of Comparative Psychology and 
other departments will be strengthened by the addition of 
collaborators. 

The attention of psychologists, physiologists and medical 
practitioners is called to the fact that this is the only journal in 
any language especially devoted to this large and important 
field of research. It is our aim to make the Journal indispen- 
sable to all interested in the structure and functions of the ner- 


336 JOURNAL OF COMPARATIVE NEUROLOGY. 


vous system from whatever point of view. They will find much 
of value in the materials published, for in addition to the recog- 
nized fact that the human nervous system can be best under- 
stood structurally by a study of its phylogeny, it is now clear 
that an understanding of the reactions of lower organisms and 
especially of the evolution of action, is necessary for an appre- 
ciation of the functional significance of the human nervous 
system. 

For morphologists the Journal will continue to be, as in. 
the past, the vade mecum in its department. And we again call 
attention to the fact that by comparative neurology we mean 
not the study of the nervous system of lower organisms alone, 
but all neurological researches carried on in the comparative 
spirit and by the comparative method. 

In addition to technical articles of a special nature, there 
will be presented critical digests, synthetic reviews and editorial 
summaries of particular topics designed to give in a non-tech- 
ical way the most important results of research in each of the 
specialties in our field. 

Hereafter the Journal will appear bi-montly, and each an- 
nual volume will contain about five hundred pages. The first 
number of the new volume we hope to have ready in February. 
Every effort will be made to secure as prompt publication of 
acceptable contributions as is consistent with the high standard 
of scientific and mechanical excellence which we propose to 
maintain. 

The subscription price will be $4.00, strictly net 
(foreign subscriptions $4.30, 18s., M. 28, 22 fr... Law 2a 
All MSS. and matter for review should be sent to the 
responsible editors. Those dealing with the structure of the 
nervous system and all business correspondence should be sent 
to the Managing Editor, C. Jupson Herrick, at Denison Uni- 
versity, Granville, Ohio; those dealing with the functions of 
the nervous system and comparative psychology may be sent 
directly to Dr. Ropext M. YERKES, Psychological Laboratory, 
Harvard University, Cambridge, Mass. 


EPRERARY NOTICES: 


The Relations of Biology and Psychology.' 


The book before us has already been reviewed by a number of 
writers and needs no introduction to the readers of this journal. The 
purpose here is simply to touch upon certain points (not dwelt upon at 
length by any of these reviewers) which are of interest at once to the 
psychologist and to the biologist. 

(1) Psychophysical Evolution. ‘The papers which are here gath- 
ered together may be treated, says the author, ‘‘as each dealing with 
a narrower question, yet as having reference to the larger problem 
which may be called psychophysical evolution—the evolution of mind 
and body together” (p. 2). This conception of psychophysical evolu- 
tion is one to which the author returns again and again throughout the 
book in a way which is stimulating or exasperating according to where 
the reader stops in in the perusal of the book. 

One at first feels that the author has struck out the true solution 
of a perplexing question, and he turns the pages expectantly until he 
shall come to the convincing presentation of this great thought. But 
as he proceeds all that he finds is the cheerful assumption that this has 
already been made as clear as is necessary—and this is the source of 
the feeling of exasperation. 

The clearest brief statement of the principle of psychophysical 
evolution is that in which the author says that ‘‘the brain not a brain 
when consciousness is not there,” and ‘‘consciousness does not, on the 
other hand, produce movement without a brain” (p. 130). This most 
promising suggestion leads the reader to the natural conclusion that 
the author has in the background a point of view which justifies what 
appears upon the surface as a rather paradoxical juxtaposition of con- 
cepts usually kept quite distinct. ‘What is this point of view? 

; We all, doubtless, today, feel that brain and consciousness are 
equally genuine and valid phases of the reality of experience; and 


1 BALDWIN, J. MARK. Development and Evolution: Including Psycho-° 
physical Evolution, Evolution by Orthoplasy, and the Theory of Genetic 
Modes. The Macmillan Co., New York, 1902. 


xxx JOURNAL OF COMPARATIVE NEUROLOGY. 


this has been stated in many metaphors and similes. We are all either 
asking the question ‘‘why the mind has a body” or why the body ever 
came to have a mind. But what we still lack is an analysis of the 
origin and meaning of the distinction—its genesis and its function. 
Why do we distinguish between brain and consciousness at all, if ulti- 
mately they are so intimately one? Just how is their difference related 
to their identity ? How did the distinction originally come to be set up, 
and what modifications has it undergone in the history of scientific 
thought ? 

(2) The Psychological and the Biological. The author's discus- 
sion of the terms ‘‘psychic” and ‘‘psychological” (Chap. I, §2) is a 
hint of such an analysis. In this section Professor BaLDwIN distin- 
guishes between the ‘‘pyschological” and the ‘‘psychic” as follows: 
‘By the psychological I mean the mental of any grade, wewed from 
the outside ; that is, viewed as a definite set or series of phenomena in 
a consciousness, recognized as facts and as ‘worth while’ as any other 
facts in nature.” ‘‘This occurrence of a psychological change in an 
animal is a fact in the same sense that the animal's process of digestion 
is’ (p. 4). ‘‘The discussion of the respective spheres of these two 
sciences turns upon a distinction of points of view. On the one hand 
the psychologist as such, and for his science, must aim at the recogni- 
tion only of the facts which are psychic or mental; that is, of such as 
are facts to the consciousness iz which they occur. ‘These alone are 
psychic, and these belong to individual psychology” (p. 5). ‘‘Psychol- 
ogy, when considered as the science of mind,—that is, looked at from 
the objective point of view,—takes cognizance of the ‘psychonomic’ ; 
but when considered as a subjective science, as interpreting its own 
data, it does not; but, on the contrary, it confines itself to the psy- 
chic” (p. 8). 

By way of criticism of this, the question at once arises whether 
there is any such thing as psychology ‘‘considered as a subjective 
science” ? Many other writers have been insisting that there is no 
‘individual psychology” in this sense; there is no science of the indi- 
vidual. From this point of view, the ‘‘psychologist as such’' is no 
scientist at all; the attempt to draw a distinction between two kinds of 
psychology in this sense proves suicidal. If the difference ‘‘turns upon 
a distinction of points of view,’’ then it does not turn upon a distinc- 
tion of contents; if it is a distinction of method only, then it is 
not a distinction of subject-matter. When we take up ‘‘the stand- 
point of the observer, that of the scientific man who _ essays 
to investigate some one else’s consciousness, or that of an animal, 


Literary Notices. XXX 


the procedure is now subject to different rules and limitations” (p. 5) 
it is true, but this is essential to any science of psychology; this is not 
another kind of psychology over and above so-called ‘‘individual psy- 
chology.” Individual psychology is not scientific psychology apart 
from this. There is no sczence of psychology which deals with the 
strictly psychic. 

Moreover, this view gets the author into difficulties when he 
comes to apply it to his doctrine of psychophysical evolution. ‘‘But 
now, and this is the essential point to remark in our present connec- 
tion, so soon as we ask the psychophysical question of genesis,—that 
of the development and evolution of mind and body taken together,— 
pursuing the biogenetic method, this limitation no longer rises. to 
trouble us. We include all psychophysical facts as such in the defini- 
tion of our science. Changes in mind and body go on together, and 
together they constitute the phenomena. Both organic and mental 
states and functions may be appealed to in our endeavor to trace the 
psychophysical series of events of such, since both are objective to the 
spectator, the scientific observer’ (p. 8). Accordingly, ‘‘with the gen- 
eral understanding now arrived at, we may take a preliminary survey 
of the field in the light of certain current hypotheses. Among these 
is what is known as ‘psychophysical parallelism’ ” (p.1o). ‘‘The prin- 
ciple of parallelism assumed, we claim once for all the right fo neglect 
the relation of the two terms, mental’ and physical, in all circumstances 
whatsoever” (p. 15). 

But how can we interchange the psychical and the physical if, 
by definition, the psychic facts are facts only ‘‘to the consciousness in 
which they occur?” The law that ‘‘for science all facts are equal’ 
does not mean that the physical and the psychical can be interchanged 
without changing the ‘‘point of view.’’ And if the author here does 
not mean the psychical by his term ‘‘mental,” then how does the dis- 
cussion become relevant to the doctrine of psychophysical parallelism ? 

What Professor BALDWIN seems to mean is that the same process 
of psychophysical evolution may be stated either as psychological or 
as biological, i. e., it may be interpreted from either of these egually 
objective points of view. But this has nothing in common with the doc- 
{rine of psychophysical parallelism. The latter, as he says elsewhere, 
is a question of ultimate philosophical interpretation, while the former 
is a question of division of labor in scientific method. Not that the 


' Note that the ambiguous term ‘‘mental,” so important at this juncture, is 
not defined. Does the author here mean “‘psychic’’ or ‘‘psychological ?”’ 


XXX JOURNAL OF CoMPARATIVE NEUROLOGY. 


question of methodology does not have an important bearing upon this 
question of the distinction between the psychical and the physical, but 
this relation can not appear as long as one member of the distinction 
is taken as fixed. 

(3) The Place of Consciousness in Evolution. The place of con- 
sciousness in evolution is the same on either the Lamarckian or Dar- 
winian view. ‘This is made possible by the author's theory of organic 
selection and social heredity. One of his reviewers, indeed, thinks 
that he has not wholly escaped the fallacy of supposing that conscious- 
ness produces causal changes in the physical world of muscles. But, 
besides the author’s disavowal of such a doctrine in a reply to this re- 
view, he distinctly says in the work before us that there is a third view 
beside the theory of automatism and the theory that consciousness is a 
vera causa (p. 121). 

But what is this third view ? Does the author here intend a func- 
tional interpretation of the relation of the psychical to the physical ? 
If there is one psychophysical system, and if consciousness is simply 
the meaning of this system when it is tensional, as contrasted with the 
state of the same system when in the relatively stable equilibrium of 
habit, then consciousness can be included in the statement of the ante- 
cedent phenomena explanatory of a voluntary movement—not indeed, 
as a distinct phenomenon, but as the statement of a continuous process 
in one of its stages. To say that the same movement could take place 
without this state of consciousness is to say that the fact that it was a 
conscious movement (i. e., had this meaning as distinct, say, from an 
habitual movement) does not make it a different movement from one 
which is not conscious. Any mark or character of the movement 
makes it a different movement. In truth, no two movements are ever 
exactly alike. Of course, you may abstract from all these differences, 
but then your judgment is an hypothetical and not a descriptive one, 
and here the aim, as the author says, is to secure a scientific in the 
sense of a descriptive statement of the facts. 

The suggestion that heredity rather than variation is the fact to be 
accounted for in evolution, ‘‘that variation is normal, and that heredity 
is acquired through the operation of natural selection restricting the 
limiting variation” (p. 230), is, then, to be put alongside of this other 
contention of the author, that consciousness, in one form, is the 
growing-point of evolution from the first. Is not this tantamount to 
saying that what we call consciousness is the variable element in devel- 
opment and evolution, that consciousness represents the shifting area 
of tension in adaptation, or, to put it from the other side, the moving 


Literary Notices. XX Xili 


equilibrium, or struggle toward equilibrium, between the forces of the 
organism and the environment ? 

(4) Mind and Body. The problem of the relation of mind and 
matter, from this standpoint, becomes chiefly a question of the logic of 
scientific method. The present writer has elsewhere (PAzlosophical 
Review, May, 1903) suggested, that this is the consistent interpretation 
of Professor BALDWIN’s chapter, in the book before us, on ‘‘Mind and 
Body.” The psychical ceases to be an entity in any sense of the term, 
even in the sense of energy. Instead of the psychical being subor- 
dinated to the concept of energy, as Professor OstwaLp contends, or 
being regarded as interchangeable because universally parallel, as Pro- 
fessor BALDWIN contends, these concepts must, in time, be recon- 
structed in terms of each other, and take their place in a scientifically 
continuous series, the terminology of which remains to be worked out, 
but of which it is the great distinction of Professor BALDWIN to have 
given a hint in his otherwise paradoxical doctrine of psychophysical 
evolution. Another hint in the same direction is the recent attempt to 
define the meaning of the psychical by Professor GEorcE H. Mean, 
in the University of Chicago Decennial Publications, Volume III, 
where the psychical is defined as the process, as contrasted with the 
content, of the experience, or, to use the terms of logic which he 
employs, the psychical is identified with the copula of the judgment. 

H. HEATH BAWDEN. 


The Psychology of Action.' 


We have in this book what we have learned always to expect from 
the pen of its distinguished author, a lucid, interesting and original 
presentation of the principles of psychology. Its originality consists 
in the successful employment of terms chosen from the sphere of 
practical life, as the leading categories and principles of division. In- 
stead of the traditional classification of the subject-matter under the 
rubrics of cognitive, affective, and conative states, we have the re- 
freshing consciousness of feeling that we follow the meaning of the 
author from the beginning without being involved in a system of tech- 
nicalities. He discusses mental life under the headings of Sensitive- 
ness, Docility, and Initiative. These terms retain the content which 
they have in ordinary life while at the same time serving the purposes 


1 Outlines of Psychology; An Elementary Treatise with Some Practical 
Applications. By JosIAH Royce, Ph. D., LL.D. Mew York, Macmillan Co., 
7903. (In the Teachers’ Professional Library series, edited by President 
NICHOLAS MURRAY BUTLER.) 


XXXIV JOURNAL OF COMPARATIVE NEUROLOGY. 


of an accurate psychological analysis. The main discussion under 
these heads is preceded by four chapters entitled respectively: ‘‘In- 
troductory Definitions and Explanations,” ‘‘The Physical Signs of the 
Presence of Mind,” ‘‘The Nervous Conditions of the Manifestation 
of Mind,” ‘‘General Features of Conscious Life.” The author then 
proceeds to discuss the three chief forms of Sensttiveness, viz: ‘‘Sen- 
sory Experience,” ‘‘Mental Imagery,” ‘‘The Feelings.” The aim 
here is to ‘‘make a summary statement of the principal kinds of states 
of consciousness that occur within the range of our psychological ex- 
perience,” considered especially with relation to the sorts of physical 
conditions upon which they depend. ocility is treated in five chap- 
ters. This is the study of the ‘‘relations that bind the consciousness 
of any moment to previous experience.” ‘The ‘‘General Law of Do- 
cility” is the law of habit which is traced through its various exempli- 
fications in ‘‘Perception and Action,” ‘‘Assimilation,” ‘‘Differentia- 
tion,” and ‘‘Imitation,” which introduces to us ‘‘The Social Aspect of 


the Higher Forms of Docility.” /ztiative is discussed in a single 
chapter entitled: ‘‘The Conditions of Mental Initiative.” This is 
followed by two concluding chapters: ‘‘Certain Varieties of Emotion- 


al and Intellectual Life,” ‘‘The Will or the Direction of Conduct.” 

In general standpoint this book may be regarded as a contribu- 
tion to what is coming to be called the functional point of view in 
psychology. This is seen in the insistence upon the integrity of ex- 
perience, in the valuable critique of the doctrine of conscious elements 
as employed by the structural psychologist, and in the use throughout 
of the biological conception of habit, and even of consciousness, as 
special developments within the life of the organism for the sake of 
enabling it to adjust itself in its changing environment. Probably the 
author did not have this last point so explicitly in his aim as might be 
inferred from the statement just made, but it is only the more signifi- 
cant if such is the case. Hints of it are scattered throughout the 
book without any more explicit statement being made than that em- 
bodied, for example, in the following sentences: ‘‘The central pro- 
cesses which our images accompany form themselves a part of our 
reaction to our environment, avd our more organized sertes of mental 
images actually form part of our conduct’ (p. 160).' ‘‘Thought is either 
action or nothing” (p. 351). Here is the gist of the functional point 
of view, that all the various forms of consciousness are special devel- 
opments zwéhzn action, and, therefore, special developments of action. 


1 Ttalics ours. 


Literary Notices. XXXV 


As this, to the present reviewer, is the most significent feature of the 
entire treatment, it may be instructive to show in what ways action is 
here made fundamental. 

(1) The author says, ‘‘Fhe single facts of sense, and the single 
movements which we make, are always related to, or, as one may say, 
are differentiations of our general orientation” (p. 147). This is con- 
nected with the ‘‘tropisms of orientation.” ‘‘The reactions of orien- 
tation are among the most fundamental phenomena of healthy life.” 
‘Our sensory experience at any moment will stand partly for our more 
general activities of orientation, and partly for our more special reac- 
tions to individual objects” (p. 143). ‘‘The special acts are always 
superposed upon the general acts.” ‘‘All our particular sensory ex- 
perience will be related, not only to our special acts, but to our gen- 
eral acts of orientation, and to those experiences which result from 
these acts.” <‘‘All such sensory experiences appear to our con- 
sciousness as facts existent within a certain primitive whole, which, 
apart from differentiation, is our experience of the general orientation 
of the entire organism” (p. 146). The author, in other words. is in- 
sisting that our motor or kinaesthetic experiences (sensations and 
images) form the very core of consciousness. The kinaesthetic sensa- 
tions supply the fundamental imagery of meaning. This is equivalent 
to saying that action is the fundamental category of experience and the 
various forms of conscious experience are special developments within 
this. Here, by the way, was an excellent opportunity to clear away 
at a stroke the whole difficulty of the relation of the psychical to the 
physical, in so far as psychology is concerned, since consciousness 
here appears simply as action passing through a tensional or recon- 
structive stage. ‘‘Tension,’’ he says on a previous page, ‘‘the mu- 
tual opposition and balancing of numerous tendencies, is absolutely 
essential to normal life.” Why should consciousness any more than 
habit be hypostasized, if both are equaliy developments of action ? 

(2) The statement which the WeBEeR-FECHNER law receives Js a 
good illustration of the tendency to interpret experience from the 
standpoint of the act, from the standpoint of the organic circuit, as 
the functional psychologists would say, rather than from the standpoint 
of any one of its contained minor activities. ‘‘The law is that in 
order that differences of sensory experience should have, in two differ- 
ent cases of comparison, the same value for our reacting consciousness, 
or should appear to be equal differences, the stimuli that are compared 
in the two different cases must differ from one another, not by the 
same absolute physical difference in their magnitude, but by the same 


XXXVI JOURNAL OF COMPARATIVE NEUROLOGY. 


relative difference’ (267-8). ‘‘The psycho-physic law appears now 
to formulate a certain limit to which the Docility of the organism in 
responding to finer differences in stimulation is subject” (270). That 
is, ‘‘the psycho-physic law is not a law directly relating to our sensa- 
tions, but is rather a law of our reactions” (272). 

(3) In chapter XII we have a most illuminating statement of the 
relations of thought to action in a discussion of the psychological and 
social functions of language. Thinking differs from naive action 
chiefly in this, that in thinking we reflect on the details of the action, 
and bring the method of the action to consciousness. ‘‘One who 
thinks makes it part of his ideal to be conscious of how he behaves in 
the presence of things. And this he does because the social compari- 
son of his acts with the acts of other people not only controls the for- 
mation of his acts, but has made his observation of his own acts an 
ideal’ (284). ‘‘The consciousness of how one performs the act” is 
the very essence of thought. The abstract idea or concept is a re- 
duced act. Take the concept or ‘‘horse” or ‘‘man.” ‘‘Whoever 
knows what a horse or man in general is, knows of some kind of act 
which it is fitting to perform in the presence of any object of the class 
in question.” ‘*The name ‘man’ or ‘horse,’ the word-image asso- 
ciated with any such subject, is itself a part of a well-known act by 
which one may react in the presence of an object in the class in ques- 
tion.- For naming objects is one way of responding to their presence” 


(286-287). ‘‘Our general ideas . . . stand, therefore, for cer- 
tain. . . attitudes.” ‘‘Our mental images of outer objects are 
never to be divorced from our reactions.”” A ‘‘general idea is a con- 


scious plan of action.” 

(4) The treatment of the relation of feeling to action supports the 
same general conception. On page 296 we read, ‘‘Like the thinking 
process in general, the reasoning process develops out of conditions 
which at the outset involve a very rich, and in fact predominant pres- 
ence of feelings and of Complex emotions. That is, reasonings have 
resulted from what were at first decidedly passionate contrasts of 
opinion.” ‘Thinking, reasoning, here would appear at the setting up 
of distinctions and the introduction of control within the primitive 
predominantly affective type of consciousness. In the chapter on the 
Feelings the author refers to the traditional view of the relation of 
feeling to thought and to action as embodying an important truth, but 
seems hesitant about adopting it. ‘‘Those who “divide mental life, in, 
the well known traditional way, into the life of cognition, the life of 
feeling, and the life of will, are accustomed to assign to the feelings a 


Literary Notices. XXXVil 


stage intermediate between the life of cognition and the life of will. 
From this point of view our cognitive consciousness first furnishes to 
us the facts. In terms of our feelings we estimate the va/ues of these 
facts for us. In view of these values our acfs are determined. ‘That 
this traditional view has a real significance cannot be questioned. But 
in the present exposition of the structure and laws of consciousness 
we are not at all closely following the lines of the traditional exposi- 
tion” (164). Now the reviewer ventures to suggest that this tradi- 
tional view, with one modification, stands more in line with the general 
method of treatment pursued in this book than does the exposition 
actually given. On another page (226), the author says that in ordi- 
nary association ‘‘ the perception is relatively instantaneous.” ‘‘ The 
present sense disturbance is at once associated with a consciousness 
due to already established motor habits” (225). This suggests the real 
significance of the traditional view—which holds that we first perceive 
something, then feel interested in it, and then act upon it—only the 
traditional view regards this as a conscious act of perception, whereas, 
in truth, this initial perceptive act is automatic, instinctive or habitual. 
It is a good illustration of one of those “attitudes” to which the 
author refers elsewhere. ‘The traditional view is true except that the 
cognition which comes first is not conscious or reflective but instinc- 
tive or intuitive. And the consciousness, which the author says ac- 
companys our acts, or takes place ‘‘ side by side with the tendencies 
to action” (164), from this point of view is rather developed within 
the action at the point where and because of the fact that this instinc- 
tive perceptive habit fails to meet the exigencies of some situation. 
Thus is evolved first an emotional consciousness and then within this, 
as we have already seen, an intellectual consciousness (in this case, 
conscious cognition) which defines and controls this emotion. From 
this standpoint feeling is just unanalyzed consciousness; it is total, 
vague, impulsive consciousness; hence the significance of the analysis 
of emotion into organic, kinaesthetic and dermal sensations. 

(5) Initiative, attention, apperception, self-activity, are all traced 
back to elemental tendencies, instincts or tropisms, operating ‘‘at times 
when the results are not immediately adaptive.” If the adaptation 
were perfectly smooth and unimpeded there would be no need for the 
evolution of such phenomena as attention and initiative. These are 
the product of, and are developed to meet the necessities of disadap- 
tation in experience. The orderly control of experience in attention 
and direction of experience in so-called self-activity are the result of a 
selection from among a great number of still unadaptive movements 


XXXViill JoURNAL OF COMPARATIVE NEUROLOGY. 


in which the animal persists despite their inefficiency. This is what 
Professor BALDWIN has called functional selection from excess move- 
ments or over-production of variations in the individual. Important 
questions arise at once to some of which an answer is given in chapter 
XIII. Why are these tropisms not immediately adaptive, and why do | 
animals persist in making these non-adaptive movements ? Why should 
such conscious processes as attention develop thus at the points of dis- 
adaptation in experience ? What is the psychology of this disadapta- 
tion or break inthe experience? And, even more important, the 
psychology of the reconstruction or readjustment after the break, by 
means of the conscious attention thus evolved. Professor ROoycE an- 
swers the first of these questions by saying that ‘‘this factor, this pecu- 
liar persistence, belongs to the temperament of the animal” (315). He, I 
suppose, would hasten to add that this is no real explanation, since 
‘‘temperament” is something itself to be explained rather than the 
- explanation of anything. Would it be in line with his own argument 
to suggest that the approximate reason is that the ordinary inhibitory 
effect of the regular routine of habitual acts is removed. The animal 
is, so to speak, reduced to a state of psychoplasm or impulse because 
of the ineffectiveness of the customary modes of activity. The rest- 
lessness and persistence in unadaptive movements represent simply 
the releasing of tendencies which are ordinarily inhibited. Relative 
freedom from ordinary restraints results in a relapse into a compara- 
tively primitive state of unmediated impulse, until new restraints can 
be established, new habits built up. This gives us a hint, at the same 
time, as to the true nature of the break or disadaptation and a sugges- 
tion as to the law of the readjustment or reconstruction. Apart from 
some such interpretation, one is impelled constantly, throughout this 
whole discussion of initiative, to ask the old question, whether there 
is ever any absolutely novel element in experience, and, if not, how 
there can be any real progress. 

(6) One’s feeling, after reading this delightful book, is one of satis- 
faction in finding the emphasis thrown once again upon the unity and 
continuity of experience, after so much analysis and dissection in re 
cent psychology, but with this, perhaps out of it, springs a desire that 
the author had carried out his organic view of experience a little further 
and shown us, not only that action is the natural consummation of feel- 
ing and thinking, but also how feeling and thinking first appear because 
of the interruption of action. It is just the full appreciation of the 
significance of this emergence of consciousness within action, as itself 
a phase of action—that consciousness not only leads over into action 


Literary Notices. XXXIX 


but arises from and within action—that is most needed at the present 
time to put psychology into right relations with biology, on the one 
side, and with philosophy, on the other. With biology, because 
herein we find a category common to both sciences—the category of 
action, of adaptation, or adjustment and readjustment. With philos- 
ophy, for a similar reason, that in the process of the reconstruction of 
experience we see the true functional significance of the psychical. 
One of the best features of the book before us is its insistance on the 
social character of consciousness, and upon the psychical individual 
as the centre for the initiation of new and progressive phases of social 
life. ‘‘Certainly a general view of the place which beings with minds 
occupy in the physical world strongly suggests that their organisms 
may especially have significance as places for the initiation of more or 
less novel types of activity” (301). ‘‘Social inventiveness depends 
upon individualistic restlessness” (327). 

A recent writer has said that ‘‘we ought to turn our views of hu- 
man psychology upside down and study what is now casually referred 
to in a chapter on habit or on the development of the will, as the gen- 
eral psychological law, of which the commonly named processes are 
derivatives.” This Professor Royce has done in a way that will prove 
instructive to psychologists as well as to teachers. 


H. HEATH BAWDEN. 


The Fore-Brain of the Bird.' 


The bird presents in its brain as in other features of its organiza- 
tion more marked specialization than is to be found in any other class 
of vertebrates. The bird brain has been the subject of comparatively 
few researches and our knowledge of its structure and of the signifi- 
cance of its several parts has been meager. In the fore-brain, especi- 
ally, great difficulties to a right understanding of its morphology and 
physiology have been presented by the unusual size of its basal gan- 
glion and the apparent absence of a true pallium over a large part of 
the fore-brain. The homology of the several parts of the basal gan- 
glion, the extent, structure, and the connections of the pallium and 
the functional significance of the several areas or nuclei are perplex- 


1 Untersuchungen iiber die vergleichende Anatomie des Gehirnes, von Dr. 
Lupwic EDINGER in Frankfurt a. M. 5. Untersuchungen tiber das Vorderhirn 
der Végel in Gemeinschaft mit Dr. A. WALLENBERG in Danzig und Dr. G. M. 
Ho.tMeEsin London. Mit sieben Tafeln und elf Textabbildungen. Sonderabd. 
aus d. Adhdlgn. d. Senckenb. naturf. Gesellschaft, BA. XX, Heft IV. 


sol JOURNAL OF COMPARATIVE NEUROLOGY. 


ing questions to the solution of which Dr. EpinGer has applied him- 
self during several years. The very satisfactory results published in this 
paper EDINGER attributes in large part to the codperation of WALLEN- 
BERG and Hotmgs, which made it possible to study a very wide range 
of material, representing all the chief types of birds, and to study the 
course of fiber tracts by the degeneration method... The attempt 
to describe the centers and fiber tracts in so complete a manner that 
they may be recognized in any group of birds, may be regarded as 
fairly successful. 

The key to the interpretation of the bird fore-brain is found in its 
development. ‘The fore-brain in the early embryo presents the typical 
arrangement: a thick ventral wall formed by the basal ganglion and an 
extensive pallium forming the roof of the wide ventricle. The basal 
ganglion, however, grows much more rapidly than the pallium and 
eventually obliterates a large part of the ventricle and fuses with the 
pallium over the greater part of the lateral and dorsal regions. The 
ventricle becomes reduced to a narrow medio-dorsal cleft connecting 
occipital and frontal horns, the latter extending into the olfactory lobe. 
The parts of the pallium thus fused with the basal ganglion have been 
overlooked or wrongly interpreted by previous authors. It is usually 
marked off from the basal ganglion by a layer of cells or by a layer of 
medullated fibers—the Stabkranz—and even where it is not so marked 
its structure and connections, as weil as its developmental history, show 
that it is a true pallium. 

In the basal ganglion the author has identified the epistriatum and 
the nucleus thaeniae and their fiber tracts, the relations being essentially 
the same as in the brain of reptiles. The remainder of the basal gan- 
glion, which is very greatly enlarged as compared with that of any 
other vertebrate, consists of a ventro-median mesostriatum, a dorsal 
hyperstriatum, and a lateral ectostriatum. ‘The ventro-anterior end of 
the mesostriatum is divided into two nuclei, a median lobus_parolfac- 
torius and a lateral nucleus basalis. Although there are great differ- 
ences in the size and functional importance of these two nuclei and of 
other parts of the basal ganglion, the brains of all birds agree in the 
main features. ‘The fibers from the pallium and the hyperstriatum 
form a medullated fiber layer, the lamina medullaris dorsalis, over the 
dorsal surface of the mesostriatum which corresponds to the capsula 
interna of mammals. The fibers then pass downward through the 
mesostriatum to form the brachia cerebri on its ventro-caudal surface. 
A true capsula interna is occasionally present (parrot). | It is as yet 


Literary Notices. xli 


impossible to compare the parts of the basal ganglion of birds with the 
nuclei in the corpus striatum of mammals. 

Only a few of the more important facts in the arrangement of the 
fiber tracts can be mentioned here. The olfactory apparatus is very 
poorly developed. Only a single tract of fibers connects the lobus 
olfactorius with the rest of the fore-brain and the destination of these 
fibers is not described. ‘The nucleus thaeniae sends a bundle to the 
ganglion habenulae. This is joined by a bundle from the occipital 
cortex (tractus cortico-habenularis) and by one from the more anterior 
portion of the basal ganglion. This bundle, which is not commented 
upon by the author, suggests the anterior portion of the tractus olfacto- 
habenularis, as it has been described in fishes. No tract which can 
certainly be considered as fornix has been found. The greater number 
of fibers, both ascending and descending, connecting the fore-brain 
with the thalamus are related to the striatum and not to the pallium. 
Especially interesting is a tract from the sensory nucleus of the V nerve 
to the nucleus basalis of the mesostriatum and a corresponding descend- 
ing tract of the oblongata and possibly to the cervical cord. 

The most of the fiber bundles connecting the pallium with other 
divisions of the brain are mingled with those of the striatum and are 
ascending fibers from the thalamus and mid-brain. Almost the only 
large descending tract from the pallium is the tractus septo-mesenceph- 
alicus, from the medio-dorsal portion of the pallium to the dorsal part 
of the thalamus and the tectum opticum. Its function is unknown. 
A commissura pallii connects the medio-dorsal cortex of the two sides. 
A tract connects the occipital cortex with the mid-brain beneath the 
tectum opticum. ‘This bundle corresponds to the cortical optic tract 
in mammals. The various portions of pallium are interconnected by 
shorter and longer associational fibers. These are least developed in 
the medio-dorsal cortex. Other important fiber tracts connect the 
nuclei of the fore-brain with one another. The anterior commissure 
appears to be purely a commissure between the two epistriata. 

The experimental works of SCHRADER, GOLTz, and KALIscHER on 
the functions of the bird fore-brain are reviewed, and the results ex- 
tended by means of the new anatomical facts. The fore-brain is not 
essential to either sensory or motor activities but exercises a directive 
influence on both which raises them above the plain of simple automa- 
tism. Removal of the pallium alone does not cause the bird to starve, | 
unless the striatum also is injured. A certain degree of localization of 
function is present in the fore-brain of various birds. The mesostria- 
tum has an important relation to the act of eating, probably mediate 


xhii JouRNAL OF COMPARATIVE NEUROLOGY. 


by the tracts connecting it with the nuclei of the V nerve in the ob- 
longata. The occipital cortex is important for sight. The median 
portion of the dorsal cortex seems to be especially concerned in the 
innervation of the limbs. 

The text of 84 quarto pages is accompanied by eleven text-figures 
and seven plates, five occupied by elegant colored figures representing 
WEIGERT sections and two presenting the results of degeneration ex- 
periments. The paper is one of the most important recent contribu- 
tions to the morphology of the vertebrate brain. 

J. B. JOHNSTON. 


The Optic Chiasma and the Post-optic Commissure. 


This subject has been treated in two papers by Dr. Burron D. 
Myers. In the first’ the chiasma alone was studied by the degener- 
ation method in the toad, cat, dog, rabbit, monkey, owl and snake, the 
toad receiving the most thorough treatment. In the toad certainly and 
probably in the owl and snake the decussation in the chiasma is total. 
In the dog, cat and monkey the decussation is unquestionably partial. 

In the second paper’ the same author makes a more thorough 
study of the relations in the rabbit, using the method of v. GUDDEN. 
The optic nerves and tracts do not begin to become medullated until 
twelve hours after birth; accordingly enucleations of the eye made 
during the first day will result in total failure of medullation of the cor- 
responding optic nerve fibers and very clear pictures can be secured by 
the WEIGERT method, the animals having been killed at various inter- 
vals after the operation. 

The experiments show conclusively that the chiasma of the rabbit 
is partial, though the uncrossed fibers are few in numher. The rela- 
tion of the optic fibers to the post-optic or inferior commissure can be 
determined by reason of the fact that the optic nerves become medul- 
lated earlier than the commissure. Comparisons of series of different 
ages made after enucleation of both eyes with similar series made after 
the enucleation of one eye permits an accurate study of the relations of 
the optic tracts to the commissure. In brief, three such commissures 
are recognized : ; 


1 The Chiasma of the Toad (Bufo lentiginosus) and Some Other Vertebrates. 
Zeits. f. Morph. u. Anthropologie, LLY ..225/ 1901. 


2 Beitrag zur Kenntniss des Chiasmas und der Commissuren am _ Boden 
des dritten Ventrikels. Archiv f. Anat. u. Physiol. Anat. Abt., 1902. 


Literary Notices. xliii 


1. The commissura inferior of GUDDEN, or Commissura arcuata 
posterior of HANNOVER.. GUDDEN termed this the com. inferior in 
transverse sections, but the com. superior MEYNERTI in horizontal sec- 
tions. Its fibers are closest to the chiasma at their crossing and go out 
laterally closely associated with the optic tracts. 

2. The rostral part of the decussatio subthalamica anterior of 
GANSER. This is GUDDEN’s commissure of MEYNERT, and it may 
properly retain the name, MEYNERT’S commissure. Its crossing lies dor- 
sally of the com. inferior, and its fibers also go out with the optic tracts 
laterally. 

3. The caudal part of the decussatio subthalamica of GANSER. 
This may retain the name GANSER’s commissure. Its fibers after the 
crossing run back on each side of the body nearer the median line than 
either of the others and they envelop the fornix tracts laterally of the 
third ventricle. 

The contradictory accounts of earlier authors are carefully re- 
viewed in the light of the author’s experimental results and it is to be 
hoped that the conclusions reached may set at rest the synonomy of 
this confusing region. It is needless to add that a real understanding 
of these commissures (or decussations, as they probably are) cannot be 
hoped for until we know the exact terminal relations of all the types of 
neurones involved. 

There is described a curious mesial slip of the optic tract which 
runs up along the inner side of the inferior commissural tract. ‘There 
is, the reviewer may add, an exactly similar detached portion of the 
optic tract in the bony fishes, which terminates in the nucleus geni- 
culatus externus. Cae he 


Peripheral Nerve Endings in Amphioxus.'’ 


The description of the course and distribution of the sensory and 
motor nerves confirms in general the results of HEyYMANS and VAN DER 
STtRIcHT. The most important part of the paper deals with the periph- 
eral endings of the sensory nerves. ‘Two sets of fibers are distin- 
guished. The first pass through the homogeneous Hautschicht by 
means of special canals and reach a position immediately beneath the 
epithelial cells. Here they branch and form a subepithelial plexus 
from which ‘‘eine Menge feinster varicéser _Fadchen” pass up and end 
between the epithelial cells, The second set of fibers have their cells 


' Das periphere Nervensystem des Amphioxus (Branchiostoma lanceola- 
tum). Von A. S. DocteL, St. Petersburg. Axat. Heften, Eletitele xevalenn pie 
147-213, Pl. XII-XXIX. 1902. 


xliv JOURNAL OF COMPARATIVE NEUROLOGY. 


of origin in the epithelium. ‘These resemble the sense cells of inverte- 
brates. Their outer ends reach to the surface but are not provided with 
sense hairs. Their inner ends are continued centrally as fibers which 
enter the sensory nerves. ‘These sense cells are found generally in the 
epithelium and especially in the oral tentacles where they are grouped 
to form the special sense organs which have heretofore been compared 
with the taste buds and end buds of typical vertebrates. Nothing is 
stated with regard to the central relations of these two sets of fibers. 
The author describes certain structures which he regards as spinal 
ganglia but their finer structure is not sufficiently well made out to 
warrant any conclusion as to their character. J. Bivde 


On the Lobus Impar of the Brain of Cyprinoid Fishes.' 


In his work entitled, ‘‘Vom Bau des Wirbeltiergehirns,” B. Hat- 
LER describes extensive anastomoses between nerve cells in the lobus 
impar of the medulla oblongata of the cyprinoid fishes and states 
that this structure is an especially favorable object for the demonstra- 
tion of such anastomoses. GROTH examined haemalum, carmine and 
GoLG! preparations of several carp-like fishes in order to check up the 
observations of HALLER, but without finding any evidence of such 
anastomoses. ‘There is an extensive but uncritical review of the litera- 
ture of these brains and some description from his own preparations of 
the anatomical structure of this part of the brain, in the course of 
which, however, nothing of morphological importance is brought out. 

Ce Je Ve 


'GrotH, A. Ueber den Lobus impar der Medulla oblongata bei Cypri- 
noiden. Dissertation. Miinchen, 1900. 


Volume XIll. APRIL, 1903. Number |. 


THE JOURNAL 


OF 


Comparative Neurology 


EDITED BY 
C. L. Herrick, ALBUQUERQUE, NEW MEXICco. 


ASBOCIATED WITH 
OLIVER S. StRoNG, Co_umBia UNIVERSITY, 
C. juvson HERRICK, DENISON UNIVERSITY. 


AND WITH THE COLLABORATION or 


LEWELLYs F. BARKER M.B., University of Chicago and Rush Medical College; FRANK 
J. Coie, University College, Liverpool; HENRY H. Donatpson, Ph.D., Unétverstty 
of Chicago; ProFessor Lupwic EnincEr, Frankfurt. a-M.; PROFESSOR A. VAN 
GEHUCHTEN, Université de Louvain; C. F. Hopcs, Ph.D., Clark Unsverstty ; 
| G. Cari Huser, M.D., University of Michigan; B. ¥. Kincsspury, Ph.D., Cor- 
nell University and the New York State Veterinary College; FREDERIC S. LRE, 
Ph.D., Columbia University; ADOLF MEYER, M.D., Pathologtcal In- 
stitute, New York; A, D. MorriLt, M.S., Hamilton College; G. A. 


540.5 


$2 2 Annual Subscription, $3.50, 


Parker, S.D., Marvard Uneversity. 


E: 
f 
a 
2 
} 


PusiisHeD py THE Epirors, DeNisON UNIVERSITY, GRANVILLE, OHIO. 


ra R. Friedlander & Son, Berlin, European Ageats, 


THE JOURNAL OF 


Comparative Neurology 


A Quarterly Periodical Devoted to the 
Comparative Study of the Nervous System in all of its Aspects. 


EDITED BY 


C. L. Herrick, Magdalena, New Mexico. 
ACS SiO'E A TAR De Wolk Ee 


OLIVER S. STRONG, Tutor in Comparative Neurology, Columbia University 
C. Jupson Herrick, Professor of Zoology, Denison University. 


AND WITH THE COLLABORATION OF 


LEWELLYs F. BARKER, M.B., Professor of Anatomy, University of Chicago and 
Rush Medical College. 


FRANK J. CoLe, Demonstrator of Zoology, University College, Liverpool. 

Henry H. DonaLpson, Ph.D., Professor of Neurology, University of Chicago. 

Prorgssor LUDWIG EDINGER, Frankfurt, a-M. 

PROFESSOR A. VAN GEHUCHTEN, Professor of Anatomy, University of Louvain, 
Belgium. 

C. F. Hopes, Ph.D., Professor of Physiology and Neurology, Clark Untversity. 


G. CARL Huser, M.D., /untor Professor of Anatomy and Director of the 
Histological Laboratory, University of Michtgan. 

B. F. Kincspury, Ph.D., Department of Physiology, Cornell University. 

FREDERIC S. LEE, Ph.D., Adjunct Professor of Physiology, Columbia Universtty. 

ApDoLF MEYER, M.D., Dérector of the Pathological Institute, New York State 
Hospitals. 

A. D. MorriLi, M.S., Professor of Biology, Hamilton College. 

G. H. ParKeEr, S.D., Assistant Professor of Zoology, Harvard University, 


THE JOURNAL OF COMPARATIVE NEUROLOGY issues one volume each year. 
Four numbers usually make a volume, though fascicles may appear at any time. 
Back numbers may be obtained at the regular rate of $3.50 per volume. Single 
numbers are also sold, the price varying with the contents of the number. 


Subscription Price $3,50 per Annum 


POST-FREE TO ALL COUNTRIES IN THE POSTAL UNION. 


Address business communications to 
C. Jupson Herrick, Manager, Denison University, Granville, Ohio, U.S, A. 


(Entered in the P. O. at Granville, Ohio, as second-class matter.) 


Printed at the University Press. 


aN COE | Oe | Oe 


- 


THE ANA-MIAS 


yield readily to organic, or true animal iron treatment, 


A resort to inorganic iron preparations or tonics, serves only to 
stimulate corpuscular proliferation without supplying sufficient nu- 
trition to mature the blood cells. 


contains 10% ANIMAL IRON, 20% coagulable albumen, and ev.zy 
element of nutrition of the animal, mineral, and vegetable kingdoms. 


BOVININE administration causes quick increase of the leucocytes, 
and a consequent arrest of all pathological processes. 


BOVININE is advertised to the Profession only. Its formula 
is open to all, 


A postal brings you our Hand-book on Haematherapy, giving 
valuable information to both the general practitioner and the specialist. 


THE BOVININE COMPANY 
75 W. HOUSTON STREET, NEW YORK 


PPO, 


Se ea 


ee 


a i al 


JOURNAL OF INEBRIETY 


PUBLISHED AT HARTFORD, CONN. 


ts the only Journal published devoted to the medical study of 
Inebriety, Alcoholism, Opium and other drug manias. 


There is no branch of medical or social science of more interest, and less understood 
by all classes. 

Every leading and professional man who would keep up with the progress of the 
times, should read this Journal. 

It comes home to every one who is interested in this topic, full of suggestions and 
instruction, pointing out the practical solution of one of the most difficult problems of 
modern times. 

Established in 1876, it was the first, and is the only Journal published, presenting 
exclusively the scientific side of this subject. 

This Journal occupies a field entirely new, and conflicts with no other publication, 
and hence has a circle of readers not confined_to any one class. 


Subscription Price, only $2.00 per Year. 


This Journal is the organ of the AMERICAN ASSOCIATION FOR THE STUDY AND CURE 
oF INEBRIETY, and is taken in all the Asylums and Hospitals in this country, and many 
in Europe. It is accepted as authority, and circulated among physicians, lawyers, judges, 
and philanthropists, from Matne to California, and in the Proyinces of Canada, and all 
the European nations. 


Address, aie By: Ca eCyl HERS. M. Be 


(EDITOR.) 
HARTFORD, CONN, 


ae 


2 


DEF ~~ 


CONTENTS. 


CONTRIBUTED ARTICLES. 


The Fore-Brain of Macacus. By Wm. Wore Lesem, M. A. 
(Columbia University.) With Plates I and II. 


Brain-Weights of Animals with Special Reference to the 
Weight of the Brain in the Macaque Monkey. By 
Epwarp Antuony Sritzka, M.D. (From the Anatomt- 
cal Laboratory, Columbia University.) 


A Description of Charts Showing the Areas of the Cross-Sec- 
tions of the Human Spinal Cord at the Level of Each 
Spinal Nerve. By Henry H. Donatpson and Davip 
J. Davis. (From the Neurological Laboratory. of the 
University of Chicago.) With Chart I. 


The Brain of the Archzoceti. By G. Exrtior Suita, M. A., 
M. D., Fellow of St. Johns College, Cambridge ; Profes- 
sor of Anatomy, Egyptian Government School of Medt- 
cine, Catro. With four figures in the text. 


LITERARY NOTICES. 


Functional Changes in the Dendrites of Cortical Neurones, i—The 
Morphological Position of the Chorda Tympani in Reptiles, 
ii—Mendel and Jacobsohn’s Jahresbericht; Fifth Issue, 
ii—Nervous System of Myxine, iii—Taste and the Fifth 
Nerve, tii—The Phylogeny of the Pallium, iii—Obsessions 
and Psychasthenia, xiv—McMurrich’s Embryology, xv—Mo- 
tor Nerve Termini in Insects, xvii—The Comparative 
Anatomy of the Brains of Lemurs and Other Mammals, xix— 
Development of Lepidosiren, xx. 


PAGE 


41 


fin Enumeration of the Medul- a 
lated Nerve Fibers in the Dorsal an 


Roots of the Spinal Nerves of Ss 
Man. | vas 


By CHARLES INGBERT, 
With Thirty:Two Figures. 


(From the Neurological Laboratory of the University of 
Chicago.) 


[Reprinted from THE JOURNAL OF COMPARATIVE 
NEUROLOGY, Vol. XIII, No. 2, 1903.] 


THE JOURNAL OF 


Comparative Neurology 


A Quarterly Periodical Devoted to the 
Comparative Study of the Nervous System in all of its Aspects. 


EDITED BY 


C. L. Herrick, Magdalena, New Mexico. 


ASS OC LATED: Wl? 


OLIVER S. STRONG, Tutor in Comparative Neurology, Columbia University 
C. JupsoN HERRICK, Professor of Zoology, Denison University. 


AND WITH THE COLLABORATION OF 


LEWELLYS F, BARKER, M.B., Professor of Anatomy, University of Chicago ana 
Rush Medical College. 


FRANK J. CoLE, Demonstrator of Zoology, University College, Liverpool. 
Henry H. Donapson, Ph.D., Professor of Neurology, University of Chicago. 
PROFESSOR LUDWIG EDINGER, Frankfurt, a-M. 


PROFESSOR A. VAN GEHUCHTEN, Professor of Anatomy, University of Louvarn, 
Beletun. 


C. F. Hopce, Ph.D., Professor of Phystology and Neurology, Clark University. 


G. CARL Huser, M.D., Junior Professor of Anatomy and Director of the 
Histological Laboratory, University of Michigan. 


R. F. Kinespury, Ph.D., Department of Physiology, Cornell University. 
FREDERIC S, LEE, Ph.D., Adjunct Professor of Phystology, Columbia University. 


ADOLF MEYER, M.D., Dérector of the Pathological Institute, New York State 
Hospitals. 


A.D. Morri1tt, M.S., Professor of Biology, Hamilton College. 
G. H. PARKER, S.D., Assistant Professor of Zoology, Harvard University. 


THE JOURNAL OF COMPARATIVE NEUROLOGY issues one volume each year. 
Four numbers usually make a volume, though fascicles may appear at any time. 
Back numbers may be obtained at the regular rate of $3.50 per volume. Single 
numbers are also sold, the price varying with the contents of the number. 


Subscription Price $3,50 per Annum 


POST-FREE TO ALL COUNTRIES IN THE POSTAL UNION. 


Address business communications to 
(*, JUDSON HERRICK, Manager, Denison University, Granville, Ohio, U.S. A. 


(Entered in the P. O. at Granville, Ohio, as second-class matter.) 


Printed at the University Press. 


7 - 
erie 
Sor 


ON THE PHYLOGENY AND MORPHOLOGICAL 
POsnON;, OF THE TERMINAL : BUDS. OF 
PISHES. 


By C, JUDSON HERRICK, 


(Studies from the Neurological Laboratory of Denison University, No. XVII.) 


[Reprinted from THE JOURNAL OF COMPARATIVE NEuROLOGY, Vol. XIII, 
No. 2, 1903.] 


i 


On the Nature of the Pericellular 
Network of Nerve Cells. 


By SHINKISHI HATAI, 


With One Plate. 


, (From the Neurological Laboratory of the University of 


Chicago.) 


{Reprinted from THE JOURNAL OF COMPARATIV 
NEvuROLOGY, Vol, XIII, No. 2, 1903.] 


ad 


Ee a WOOO NUACE ©: .O-F 


Comparative Neurology 


A Quarterly Periodical Devoted to the 
Comparative Study of the Nervous System in all of its Aspects. 


EDITED. BY 


C. L. Herrick, Magdalena, New Mexico. 


AS3 OG DAT ED Wel TH 


OLIVER S. STRONG, Tutor in Comparative Neurology, Columbia University 
C. Jupson HERRICK, Professor of Zoology, Denison University. 


AND WITH THE COLLABORATION OF 


LEWELLYS F. BARKER, M.B., Professor of Anatomy, University of Chicago and 
Rush Medical College. 


FRANK J. CoLe, Demonstrator of Zoology, University College, Liverpool. 
Henry H. DonaLpson, Ph.D., Professor of Neurology, University of Chicago. 
PROFESSOR LUDWIG EDINGER, Frankfurt, a-M, 


PROFESSOR A. VAN GEHUCHTEN, Professor of Anatomy, University of Louvain, 
Belgtum. 


C. F. Hopce, Ph.D., Professor of Physiology and Neurology, Clark University. 


G. CARL Huser, M.D., Junior Professor of Anatemy and Directer of the 
Histologwal Laboratory, University of Michigan. 


B. F. Kincspury, Ph.D., Department of Physiology, Cornell University. 
FREDERIC S. LEE, Ph.D.. Adjunct Professor of Physiology, Columbia University. 


ADOLF MEYER, M.D., Director of the Pathological Institute, New York State 
Hospitals. 


A.D. MorriLu, M.S., Professor of Biology, Hamilton College. 
G. H. PARKER, S.D., Assistant Professor of Zoology, Harvard University. 


THE JOURNAL OF COMPARATIVE NEUROLOGY issues one volume each year. 
Four numbers usually make a volume, though fascicles may appear at any time. 
Back numbers may be obtained at the regular rate of $3.50 per volume. Single 
numbers are also sold, the price varying with the contents of the number. 


Subscription Price $3,50 per Annum 


POST-FREE TO ALL COUNTRIES IN THE POSTAL UNION. 


Address business communications to 
C. JupsoN HERRICK, Manager, Denison University, Granville, Ohio, U.S. A. 


(Entered in the P. O. at Granville, Ohio, as second-class matter.) 


Printed at the Universtty Press. 


The Neurokeratin in the Medul- 
lary Sheaths of the Peripheral 
Nerves of Mammals, 


By SHINKISHI HATAI, 
With One Plate. 


(From the Neurological Laboratory of the University of 
Chicago.) 


{Reprinted from THE JOURNAL OF COMPARATIVE 
NEvUROLOGY, Vol. XIII, No. 2, 1903.] 


WH Ber RIN AL. OF 


Comparative Neurology - 


A Quarterly Periodical Devoted to the 
Comparative Study of the Nervous System in all of its Aspects. 


EDITED BY 


C. L. Herrick, Magdalena, New Mexico. 


ALS. 9) OC GAVE EDS WLer Ee 


OLIVER S. STRONG, Tutor in Comparative Neurology, Columbia University 
C. Jupson HeErRrRIcK, Professor of Zoology, Dentson University. 


AND WITH THE COLLABORATION OF 


LEWELLYs F, BARKER, M.B., Professor of Anatomy, University of Chicago and 
Rush Medical College. 


FRANK J. CoLE, Demonstrator of Zoology, University College, Liverpool. 
Henry H. DoNALDSON, Ph.D., Professor of Neurology, University of Chicago. 
PROFESSOR LUDWIG EDINGER, Frankfurt, a-M, 


PROFESSOR A. VAN GEHUCHTEN, Professor of Anatomy, University of Louvain, 
Belgium. 


C. F. Hones, Ph.D., Professor of Physiology and Neurology, Clark University. 


G. CarL Huser, M.D., /untor Professor of Anatomy and Director of the 
Fiistological Laboratory, University of Michzgan. 


B. F. Kinespury, Ph.D., Department of Physiology, Cornell University. 

FREDERIC S. LEE, Ph.D., Adjunct Professor of Physiology, Columbia University. 

ADOLF MEYER, M.D., Director of the Pathological Institute, New York State 
~ Hospitals. 

A.D. Morriti, M.S., Professor of Biology, Hamilton College. 

G. H. PARKER, S.D., Assistant Professor of Zoology, Harvard University. 


THE JOURNAL OF COMPARATIVE NEUROLOGY issues one volume each year. 
Four numbers usually make a volume, though fascicles may appear at any time. 
Back numbers may be obtained at the regular rate of $3.50 per volume. Single 
numbers are also sold, the price varying with the contents of the number. 


Subscription Price $3,50 per Annum 


POST-FREE TO ALL COUNTRIES IN THE POSTAL UNION. 


Address business communications to 
©, Jupson HERRICK, Manager, Denison University, Granville, Ohio, U.S. A. 


(Entered in the P. O. at Granville, Ohio, as second-class matter.) 


Printed at the University Press. 


Volume XiIll. OCTOBER, 1903. Number 3. 


PE JOURNAL 


OF 


Comparative Neurology 


EDITED BY 
C. L. Herrick, MacpALena, New Mexico. 


ASSOCIATED WITH 
OLIVER S. StRonG, COLUMBIA UNIVERSITY, 
C. Jupson Herrick, DENISON UNIVERSITY. 


AND WITH THE COLLABORATION OF 


LEWELLYs F, BARKER M.B., University of Chicago and Rush Medical College; FRANK 
. J. Cote, Ontversity College, Liverpool; HENRY H, DoNALDSON, Ph.D., University 
of Chicago; PRoFEssor Lupwic EDINGER, /rankfurt. a-M.; PROFESSOR A. VAN 
GEHUCHTEN, Université de Louvain; C. F. Hopcr, Ph.D., Clark University ; 
G, Cart Huser, M.D., University of Michigan; B, F. Kincspury, Ph.D., Cor- 
nell University and the New York State Veterinary College; FREDERIC S. LEE, 
Ph.D., Columbia University; ADOLF MEYER, M.D., Pathological In- 
stitute, New York; A. D. Morritit, M.S., Hamelton College; G. Hi. 
PARKER, S.D., Harvard University. 


Annual Subscription, $3.50. 


p 


PuBLISHED BY THE Epirors, DENISON UNIVERSITY, GRANVILLE, OHIO. 


R. Friedlander & Son, Berlin, European Ageats. 


THE 


JOURNAL OF COMPARATIVE NEUROLOGY 


A QUARTERLY PERIODICAL DEVOTED TO THE 


Comparative Study of the Nervous System in all of Its Aspects. 


One volume is issued each year and usually comprises four numbers, though 
fascicles may appear at any time. Back numbers may be obtained ai the 
regular rate of $3 50 per volume. Single numbers are also sold, the 
price varying with the contents. Complete tables of contents of 
ali the voiumes will be set upon application to the manager. 


Subscription Price, ; A ; ‘$3.50 per annum, 
POST-FREE TO ALL COUNTRIES IN TIE POSTAL UNION. 


Address al! business communications to 


C. Jupson Herrick, Manager, Denison University, Granville, Ohio, U.S.A. 


eeeeeseaceseseeoe 
BOC ROORIAG ; 


Ri, ca IO SN SS SDA EA Se Sa PR AR AE ER I BE, BS 
SVIOUSVEY 
a 


I ANAM LA 
and converted jato. living force. w 
3. Because Axcmia results from a form of malnutrition 
compounded pabataim that calls for no chemico-vital effort whatever. 
Apoly the crucial test—ciinical 1se—at our expense, and convince 


BOVININE overcomes Anemia logically, rationally and 
radically, for several substantial reasons: 
1. Because it supplies the starving organism with the 
requisites for immediate reparation. 
2. Because it needs no preparation or transformation at 
caused by the absolute inertia of tne digestive function. 
yourself that our claims are neither extravagant nor exaggerated, but 


SUBIC VUE 


DE ES Sig I A fs 
iter 


PL OLN PON ONIN LVL LLL NN VL ee ee 
aang eijanir Sega sa NN * “ eet 
the hands of the vital machinery before it can be assimilated 
TROWININE cories to the rescrie by supplying a vitalized and perfectly 
are strictly based cn science. 


the Bovinine Company 


3 

Qo hee 

3 75 West Houston Strect, NEW YORHA 
b ama HARE EEE DARA BAR MABBOVOHLI CK Vsore OBE SEGOHSCOCOHHOHLOSO 


THE AMERICAN NATURALIST. 


A Monthly Journal Devoted to the Natural Sciences in 
Their Widest Sense. 


Since the foundation in 1867 by four of the pupils of Louis 
Agassiz, THE AMERICAN NATURALIST has been a 
representative American magazine of Natural History and has 


played an important part inthe advancement of science in 
this country. 


The journal aims to present to its readers the leading facts 
and discoveries in the fields Anthropology, General Biology, 
Zoology, Botany, Paleontology, Geology and Mineralogy, 
and the various sub-divisions of those subjects. 


Annual Subscription, $4.00 net, in advance. Single Copies, 35c. 
Foreign Subscriptions, $4.60. 


GINNw. & COMPANY, Publishers. 


29 BEACON STEET, BOSTON, MASS. 


JOURNAL OF INEBRIETY 


PUBLISHED AT HARTFORD, CONN. 


Is the only Journal. published devoted to the medical, study of 


Inebriety, Alcoholism, Opium and other drug manias. 


There is no branch of medical or social science of more interest, and less understood 
by all classes. 

Every leading and professional man who would keep up with the progress of the 
times, should read this Journal. 

It comes home to every one who is interested in this topic, full of suggestions and 
instruction. pointing out the practical solution of one of the most difficult problems of 
modern times. 

' ¥stablished in 1876, it was the first, and is the only Journal published, presenting 
exclusively the scientific side of this subject. ‘ 

This Journal occupies a field entirely new, and conflicts with no other publication, 
and hence has a circle of readers not confined to any one class. 


Subscription Price, only $2.00 per Year. 


This Journal is the organ of the AMERICAN ASSOCIATION FOR THE STUDY AND CURE 

oF INEBRIETY, and is taken in all the Asylums and Hospitals in this country, and many 

in Eurvpe. It is accepted as authority, and circulated among physicians, lawyers, judges, 

and philanthropists, from Mazne to California, and in the Provinces of Canada, and all 
the European nations, 


Maas toy GROTHERS) My, D,, 
(EDITOR.) 
HARTFORD, CONN. 


CONTENTS, 


CONTRIBUTED ARTICLES. 


The Neurofibrillar Structures in the Ganglia of the Leech 
and Crayfish with Especial Reference to the Neurone 
Theory. By C. W. Prentiss. Parker Fellow in 
Zoilogy, Harvard University. With Plates V and VI 


On the Increase in the Number of Medullated Nerve Fibers 
in the Ventral Roots of the Spinal Nerves of the 
Growing White Rat. By SuinxisHi Hata. (from 
the Neurological Laboratory of the Untversity of Chicago) 


On the Medullated Nerve Fibers Crossing the Site of Le- 
sions in the Brain of the White Rat. By S. WALTER 
Ranson, Jnstructor in Anatomy, Marion-Sims-Beau- 
mont Medical School, St. Louis University. (From the 
Neurological Laboratory of the University of Chicago. 
With Plate VII : k 


On the Density of the Cutaneous Innervation in Man. By 
CHARLES E, IncBert. (From the Neurological Labor- 
atory of the University of Chicago.) 


Ona Law Determining the Number of Medullated Nerve 
Fibers Innervating the Thigh, Shank and Foot of the 
Frog—Rana virescens. By Henry H. DoNALpson. 
(From the Neurological Laboratory of the University of 
Chicago.) ‘ ‘ é d ‘ : ; 


157 


177 


185, 


209 


223 


Volume XI. - DECEMBER, 1903. Number 4. 


Poe JOURNAL 


OF 


Comparative Neurology 


P EDITED BY 
C. L. Herrick, Socorro, New Mexico. 


, ASSOCIATED WITH 
OLIVER S. Srronc, COLUMBIA UNIVERSITY, 
C. Jupson Herrick, Denison UNIVERSITY. 


AND WITH THE COLLABORATION OF 


LEWELLys F. BARKER, M.B., University of Chicago and Rush Medical College; FRANK 
J. Core, University of Liverpool; WENRY H. Donarpson, Ph. D., University 
of Chicago; PRoFessor Lupwic EDINGER, Frankfurt. a-M,.; PROFESSOR A. VAN 
GEHUCHTEN, Université de Louvain; C. F. Hopce, Ph.D., Clark Universtty ; 
G. Cart Huser, M.D., University of Michigan; B. ¥. Kincspury, Ph.D., Cor- 
nell University and the New York State Veterinary College; FREDERIC S. LEE, 
Ph.D., Columbia University; ApotF Mryer, M.D., Pathological In- 
stitute, New York; A. D. Morritt, M.S., Hamilton College; G. A. 
PARKER, S.D., Harvard University. 


Published by the Editors, Denison University, Granville, Ohio. 


Issued January, 1904. 


JOURNAL OF COMPARATIVE NEUROLOGY AND PSYCHOLOGY. 


The enlargement of scope and editorial facilities of Zhe Journal of 
Comparative Neurology announced on the Editorial page of this issue 
has led to a change in the name of the publication, as indicated in the caption 
of this notice, the change to take effect with the first issue for 1904. The 
Board of Editors remains as before, save that Dr. RoBERT M. YERKES, of 
the Harvard Psychological Laboratory, becomes the responsible editor for the 
Department of Animal Rehavior and the staff of Collaborating Editors is con- 
siderably enlarged. . 

Collaborators 


PRroFessor J. Mark Batpwin, Johns Hopkins University. 
Prorrssor Lewe.tys F. Barker, The University of Chicago. 
Proressor H. HeatH Bawpen, Vassar College. 

Dr. A. BETHE, Strassburg. 

Proressor G. E. Cocuitt, Pacific University, Oregon. 

Dr. F. J. Cote, University of Liverpool. 

Proressor H. E. Crampron, Columbia University. 

PROFESSOR CHARLES B. DavENPORT, The University of Chicago. 
Mr. Wm. Harper Davis, Columbia University. 

ProFessoR Henry H. Donatpson, The University of Chicago. 
PROFESSOR Lupwic EpINGER, Frankfurt a-M. 

Dr. S. I. Franz, Dartmouth College. 

PRoFEssOR A. vAN GEHUCHTEN, University of Louvain. 
PROFESSOR Ross GRANVILLE Harrison, Johns Hopkins University. 
Proressor C. F. Hopce, Clark University. 

Dr. S. I. Hotmes, The University of Michigan. 

Dr. Ep. B. Horr, Harvard University. 

Proressor G. Cart Huser, The University of Michigan. 
ProrrssoR Herpert S. JENNINGS, The University of Pennsylvania. 
PRoFEssOR J. B. JoHNston, West Virginia University. 
Proressor B. F. Kincsspury, Cornell University. 

PROFESSOR FREDERIC S. LEE, Columbia University. 

PROFEssor JacguEs Loes, The University of California. 

Dr. Apotr Meyer, Pathological Institute of the New York State Hospitals. 
PRoressoR WesLry Mitts, McGill University. 

PrincipaL C. Lroyp Morcan, University College, Bristol. 
Prorrssor T. H. Morcan, Bryn Mawr College. 

ProressoR A. D. Morritt, Hamilton College. 

PROFESSOR Huco MinsTerBerG, Harvard University. 

Proressor W. A. NaGet, Berlin. 

Proressor G. H. Parker, Harvard University. 

Dr. Stewart Paton, Johns Hopkins University. 

Dr. Raymond PEARL, The University of Michigan. 

Dr. C. W. Prentiss, Western Reserve University. 

Proressor C. S. SHERINGTON, University of Liverpool. 
PRoressor Epwarp L. THORNDIKE, Columbia University. 

Dr. Joun B. Warson, The University of Chicago. 

PRoFrEssOR Wm. Morton WHEELER, American Museum of Natural History. 
Proressor C. O. WHITMAN, University of Chicago. 


Subscription Price $4.00 a Year. (Foreign Subscription $4.30) post-paid. 


Address all business correspondence to the Manager, 
Cc. JUDSON HERRICK, Denison University, Granville, Ohio. 


AVEDA VENA DERA BADER A PENA BER ADEM ADEM AERA DER ABER ADEM AD 


The American Naturalist 


A Monthly Journal Devoted to the Natural Sciences 
in Their Widest Sense 


The journal aims to present to its readers the leading facts 
and discoveries in the fields of Anthropology, General Biology, 
Zodlogy, Botany, Paleontology, Geology and Mineralogy, and 
the various sub-divisions of those subjects. 


CONTENTS OF THE LAST NUMBER. ) 
I. Adaptations to Aquatic, Arboreal, Fossorial and Cursorial Habits 
in Mammals. II. Arboreal Adaptations. . Louis I. Dublin 5 
II. Mutation in Plants 5 : . : Dr. D. T. MacDougal 
III. Distribution of the Fresh-Water Fishes of Mexico. : : 
: ! j . : “ - : : Dr. S. E. Meek 
IV. Examination of Organic Remains in Postglacial Deposits 
: ; : : j ‘ ; ; Dr. P. Olsson-Seffer 
V. Notes and Literature: Exploration, Hatcher’s Narrative of the 
Princeton Patagonia Expedition—Zoology, A Summary of the 
Coccide, Another Text-book of Entomology, Two Papers on 
Insect Wings—Boftany, A New Flora of the Southeastern United 
States, Notes. 
All who are interested in the Scientific questions of the 
day will find this Magazine invaluable. New subscrip- ) 
é tions for 1904 are now being received. 
. GINN @ COMPANY, Publishers, ; 
¢ Boston New York Chicago London 
, San Francisco Atlanta Dallas Columbus 
( 
Car~veruser y y Y Baten 


SURGICAL SHOCK 


is reduced to a minimum by high rectal injections of BOVININE 
combined with an equal quantity of salt solution, heated to 70°F, 
and administered prior to, during, and subsequent to operation. 


ho fee |= woe a) 6f ae eee 


improves the heart action and circulation at once. 

The blood which has become non-zrated through ether adminis- 
tration is oxygenated by the introduction of a fresh supply, and is 
rapidly restored to normal condition, > To this fact is due the power 
of BOYZINE to prevent the thirst, nausea, and emesis, which 
usually follow anzsthesia. 

Reports of numerous cases are cited in our Scientific treatise on 
Hzmatherapy. It is yours for the asking. 


he Bovinine Company 
73 West Houston Street, NEW YORK 


CONTENTS. 


PS 
CONTRIBUTED ARTICLES. 


PAGE 


The Rate of the Nervous Impulse in the Ventral Nerve-Cord of 
Certain Worms. By O. P. Jenkins and A. J. CARLSON. 
(From the Hopkins’ Seaside Laboratory and the Phystological 
Laboratory of Leland Stanford, Jr., University.) ~With 


fourteen figures. oh Wr Ong 1 4 UE ae ae 


Notes on the Technique of Weigert's Method for Staining 
Medullated Nerve Fibers. By O. S. Strone, Coluwnbia 
Oniversity, New York’... 0. 3-23 3s 


The Doctrine of Nerve Components and Some of Its Applica- — 
fions,. By C. Jupson. Herrick. : >.) ).ee3 ae 


Columella Auris and Nervus Facialis in the Urodela. By B. | 
Py KRINGSBURY.* 


Editorial Announcement. . : : : : : ‘ : S55 


LITERARY NOTICES. 


The Relations of Biology and Psychology, xxix—The Psychology of 
Action, xxxiii—The Fore-Brain of the Bird, xxxix—The Optic 
Chiasma and the Post-Optic Commissure, xlii—Peripheral 
Nerve Endings in Amphioxus, xliii—On the Lobus Impar of 
the Brain of Cyprinoid Fishes, xliy. 


—— 


L 
! 


sii 


eee 
theie 
bbiay 
a 
tte 
ving 


the 
: Hida, ye 
’ 
pileiare bd blery 
ae herb be 
aie 
ire Seok : ° 
ba Ort tet 
‘ 
i i fea 4% 
Neve bale ‘ 
A bibs Whi 


Caters 
Atte 


tian bie 
hee 


peror' ) 
teed ‘ 
ith eats i 
efahsd,t 
oy 
/4it 


pete 


ae te 
ity 
it} 


. 
‘ 3435219 
ee 


pEge 


ue 
Ti ty 


it TE 
raitetss iiss 
= 
aa 
+