BLOOD-PARASITES OF FISHES OCCUREING AT ROVIGNO. 119 



always larger and much blacker than the chromatic grains of 

 the nucleus. The granules are of about the same size whether 

 two are present or one. They are always distinctly outside 

 the nucleus, lying usually close to it, however, about the 

 middle of the body. They were first noticed in the iron- 

 haeraatoxylin preparations, and we surmised at first that they 

 might correspond to part or all of the conspicuous vacuolar 

 region seen in Griemsa smears, but have found since that this 

 is not the case. In some individuals both grains and vacuole 

 are seen to be present (figs. 1, 13, 16, 40). Sometimes one of 

 the grains appears to lie in the vacuole, but we think that, 

 in such a case, the grain is really outside the vacuole, lying- 

 above or below it instead of at the side (cf. especially fig. 43, 

 Avhere oue grain is at the side, the other apparently in the 

 vacuole). Although the grains (or single grain) usually lie 

 close to the vacuole, this is not always so; for instance, in 

 the parasite drawn in fig. 46 the grain is well removed from 

 the nuclear zone, while in fig. 39 it is on the opposite side of 

 the nucleus, near the other end of the body. 



Turning next to Giemsa smears for indications of these 

 grains, we have found that there is often considerable difficulty 

 in i"ecognising them with certainty. This is chiefly because 

 none of the individuals show any signs of bodies which have 

 taken up the Giemsa stain in the same intense manner in 

 which the above-described granules stain with iron-hsema- 

 toxylin. In individuals which do show granules that can be 

 reasonably identified with tliose, the granules are markedly 

 smaller in size. Hence care has to be taken not to be misled 

 by stray, reddish-staining grains of the ordinary chromatoid 

 character, of which occasionally one or two are present in 

 the cytoplasm. Making all possible allowance for such, we 

 do nevertheless find in some parasites one or two definite 

 granules, situated close to the cytoplasmic vacuole, in a posi- 

 tion similar to that often occupied by the granules in the iron- 

 hsematoxylin smears, which there can be little doubt actually 

 correspond to those bodies. They stain dark-reddish in most 

 smears, about the same colour as the chromatin-masses of 



