Cutaneous Glands in the Terrestrial Tsopoda. 219 
The gland-cell is now a typical rosette (PI. II. fig. 2), 
aud in section assumes a variety of shapes (PI. II. fig. 7 
& Pl. III. fig. 8). 
Ihave nothing to add to the description of the proto- 
plasm and nucleus of Ide, excepting that in those cells 
containing vacuoles these latter may be fairly small 
(Pl. ILI. fig. 9) or large spaces (PI. III. fig. 10) filled with 
a non-granular secretion. 
So far as my observations go, I am unable to agree with or 
corroborate Ide’s account of the nuclei. His third nucleus 
may be the same as that here figured, but his fourth nucleus 
is, I my sf ures one of the nuclei of the excretory duct 
(Pl. IIL. fig. 12). In darkly stained specimens these can be 
seen from the commencement of the duct and traced at 
irregular intervals down to within a short distance of the 
external opening. 
‘In neither P. dilatatus or P. scaber could any sieve- 
plates, as described by Huet, be found on the dorsal surface 
of the pleural plates. On the other hand, a very definite 
series of pores could be distinctly made out on the antero- 
lateral border of the pleural plates in both of the above- 
meutioned species (Pl. ILI. fig. 13), and the secretion could 
be drawn out, after being placed in dilute alcohol, in 
long thread-like strands. 
It is exceedingly difficult to demonstrate these openings 
microscopically, but in pleura from which the under half 
had been removed, and then treated with formic acid, 
distinct pores could be clearly made out. 
In the process of mounting and examining the pleural 
plates one frequently sees oval or circular areas having a 
sieve-like appearance, but these are due to physical causes, 
such as the mixing of a tiny drop of alcohol and xylol, 
the chemical action of acids, ete. 
Pleural plates of various species were also treated by 
the hydrochloric acid and permanganate of soda method (1) 
and various other methods with similar results. I am there- 
fore of opinion’ that the sieve-plates of Huet do not exist, 
but that there are cutaneous glands in the mesosomatic 
segments there cannot: be the slightest doubt. In P. dila- 
tutus these glands are repeated in each segmeut of the 
mesosome, and in segments 4 and 5 (possibly also in sezment 
3) of the metasome, and also in the uropoda. 
FuNCTION. 
Like many other glandular structures in the Crustacea, 
15* 
