292 
on the nervous system of other annelids. Later, in looking over the 
plates which accompany v. LENHOSSEK’s (95) article on “Centrosom 
und Sphäre in den Spinalganglienzellen des Frosches”, I was impressed 
by the very strong resemblance which the ganglion cells of the frog 
pictured by v. LenHosséx bore to those which I had found in this 
annelid. After reading his paper there remained not the slightest 
doubt that I was dealing with the same structure which he there 
figures under the name of ““Centrosom und Sphäre”. As v. LENHOSSEK 
says that a centrosome had never before been noted in a nerve cell 
and as my material is so far removed systematically from the frog, 
a description of the cells in this annelid seems to be desirable. 
When my brief account was nearly finished a paper by DEHLER 
(95) concerning the finer structure of the sympathetic ganglion cell 
of the frog was received. In these sympathetic ganglion cells DEHLER 
finds and describes the same structure which v. LennosséK had noted 
for the spinal ganglion cells. DEHLER says that he has been the 
first to confirm v. LENHOSSEK’s discovery, and remarks that, as 
v. LENHOSSER had prophesied, the first proof of the presence of a 
centrosome in a nerve cell has received quick confirmation. I believe 
that these are the only authors who have ever described such a 
structure for any nerve cell, and consequently that no such nerve 
celis have ever before been described for an invertebrate. 
Both v. LEenHossek and DEHLER ascribe very great importance 
to the methods which they employed. Both obtained their best results 
by fixing with corrosive sublimate and staining with iron haematoxylin 
followed by bordeaux, and both state that treatment with osmic acid 
gives unsatisfactory results. DEHLER says: “Ich halte für das Studium 
der Nervenzelle die Fixirung mit Osmiumsäurelösung und ihren Com- 
binationen für weniger brauchbar, da sich die feineren Gebilde diffus, 
verschwommen, eventuell schichtweise ungleich färben”. In regard to 
the use of osmic acid, my experience differs from that of both writers, 
as I obtained my best and clearest preparations from material prepared 
by one of vom Rarn’s (93, p. 102) osmic mixtures. After reading 
v. LEnHOoss£rX’s paper, I took pains to confirm my results by his 
methods, viz. the use of corrosive sublimate followed by iron haemato- 
xylin. Since I obtained satisfactory results with the haematoxylin alone, 
I did not follow it by the use of bordeaux, as both v. LENHOSSER 
and DEHLER recommend. The conditions obtained by this method 
furnished an excellent confirmation of the preparations made with the 
vom RATH mixture, but were in no respect better. 
In the first method employed the material was allowed to remain 
