H. Henry 
193 
Type 3. (PI. VIII, figs. 21 to 30.) 
These are 8 /a to 9’4 /i long, the average length being 8'7 /a, and S o yu, 
to 4 /A wide, the average width being 3'6 /u.. Most of these large broad 
forms would appear to be adult specimens of many of the forms in type 
1. It is possible too that some of them may result from the long thin 
forms by a simple increase in width. The protoplasm is paler than 
that in type 2, with Leishman and Giemsa staining being more 
distinctly pale blue, and with none of the purple tint so frequently 
found in the latter. -The nucleus is larger, but relatively not more so 
than that in type 2, when one takes into consideration the increase in 
bulk of the whole parasite. The nuclear chromatin is more coarsely 
granular, and at times takes on the appearance of deeply staining rods, 
which zigzag across the parasite transversely (fig. 27). The pole 
opposite the nucleus is still occupied by the deeply staining, purplish 
black material found in other forms; but, on its inner aspect this 
substance does not end so abruptly as in specimens of type 2, and very 
frequently there lie actually in or in close proximity to this inner 
margin one or two clear round vacuoles (figs. 23, 27—30). Also there 
are to be found sometimes on the nuclear side of these vacuoles one or 
two deeply staining granules (figs. 27, 29). It is impossible to decide 
whether these are chromatin elements or whether they represent small 
pieces of material which have become separated from the main polar 
mass. 
Comparison with known forms. 
This haemogregarine of the catfish is quite distinct from the more 
commonly occurring haemogregarines met with in seafish. It differs 
markedly from the schizohaemogregarines, viz. Haemogregarina higemina 
of the common shanny, Blennius pholis (Laveran and Mesnil, 1901), 
Haemogregarina quadrigemina of the dragonet, Gallionymus lyra 
(Brumpt and Lebailly, 1904) and Haemogregarina simondi of the 
ordinary sole (Solea vulgaris, Laveran and Mesnil, 1901) for it shows 
no evidence of intracorpuscular fission. Moreover, the nucleus throughout 
is larger, being composed of bigger granules, and stains much more 
deeply than do those of the schizohaemogregarines treated under 
similar conditions as regards fixation and staining. Some of the long 
thin forms, and of the forms intermediate between types 2 and 3, 
resemble in size and contour the haemogregarine (Lebailly, 1904) met 
with in Pleuronectes platessa, and that occurring in Coitus bubalis 
(Brumpt and Lebailly, 1904), but in these latter the nucleus is more 
Parasitology v 13 
