W. H. Leigh-Sharpe 
281 
independently one from another without arrangement, but are collected 
into close bundles which, as well in number as in position, are, in most 
cases, constant in the same species. They can generally be distinguished 
a little in front of the anus. Afterwards the bundles increase in breadth 
through the addition of more ducts, until in the neighbourhood of the 
clitellum they are divided up into several, and in the clitellum into 
entirely separate bundles with single ducts. 
The investigation of these cells was undertaken and first adequately 
described by Johansson in Abranchus brunneus, and afterwards they 
were independently examined by Dr Henry, who did not publish any 
account, with Mr Evans, in the leech from Anarrhichas lupus which 
I am assuming throughout is Platybdella anarrhichae. Subsequently I 
Fig. 5. Platybdella anarrhichae. Longitudinal section very highly magnified showing 
how the efferent ducts of the cocoon gland cells (eff.d.) run with the longitudinal 
muscle fibres ( m .). 
have re-examined the whole matter, and am able to confirm their work 
as I have described in the foregoing account. I must however mention 
that Johansson considers the secretion to be at first completely homo¬ 
geneous, but I can easily trace the course of the ducts among the muscles 
by means of their clear granular contents as shown in Fig. 5. These clear 
granules are already plainly visible in the cytoplasm of the gland cell at 
the beginmng (Fig. 3), so cannot be deposited or precipitated in the 
secretion later. Johansson mentions that the number and position of the 
bundles in every species of Ichthyobdellid leech that he has examined 
(which does not appear to include Platybdella anarrhichae) is constant 
with two exceptions, Piscicola and Abranchus brunneus, and all the species 
show a complete similarity in this respect. The normal condition 
