1898-99. | ON THE CYTOLOGY OF NON-NUCLEATED ORGANISMS. 467 
chloric acid solution of 0.5 per cent. strength, and in very dilute nitric 
acid and sulphuric acid, while those of the first type are unaffected. As 
a rule they have no special affinity for iodine over that of the cytoplasm 
of the peripheral zone. When material, hardened in alcohol, was treated 
with the nitric-molybdate reagent for several hours and then with a 
solution of phenylhydrazin hydrochloride, these granules gave no reaction 
and were consequently not visible, while in the granules of the first type 
the outer coat, which stains deeply with hematoxylin, gave a strong re- 
action for phosphorus. As this reaction required time to develop, the 
phosphorus demonstrated must be present in organic form, that is, in 
the form which characterizes the nucleins and nucleo-proteids. 
When material, hardened in alcohol, was treated at 35° C. with sul- 
phuric acid alcohol to set free organic iron, and the iron, thus set free, 
demonstrated by either the Prussian-blue or the haematoxylin method, 
one found a distinct reaction in the granules of the first type but not in 
the second. When the granules were large, as represented in Figs. 
I1-14, the outer coat only gave the reaction, but when a granule was 
very minute the iron appeared to be uniformly distributed through it. 
The iron reaction was also obtained in these granules by placing the 
broken-up trichomes on a slide in a drop of the glycerine-ammonium 
sulphide mixture and covering the preparation with a cover glass, after 
which it is placed in a warm oven kept at a temperature of 60° C. for a 
week, during which the preparation was examined from time to time. 
The only obstacle to success in these preparations is the presence of the 
thick membrane or sheath of the trichome which, as already pointed out, 
prevents a free diffusion of the sulphide into the cells which may conse- 
quently not exhibit any reaction, but, in those portions of the trichomes 
freed from the investing sheath, the reaction comes out quite distinctly 
in from three to five days. The threads of MWicrocoleus terrestris are 
most readily freed from their membranes and if granules of the first type 
are present, which is the case ifthe form is growing vigorously, they give 
the dark-green reaction of ferrous sulphide in about three days, and it is 
distributed as it is found after treatment with acid alcohol, that is, in the 
outer portions of each granule. No reaction was obtained in granules of 
the second variety.’ 
The iron demonstrated in this case can belong only to the “ masked ” 
form, and as such is most usually associated with a nuclein or nucleo- 
proteid. This and the fact that the granules in question contain organic 
/ 
x An exception must be made in the case of those preparations referred to in my former paper, “‘ Quart. 
Journ, Micro. Sci.,” Vol. XXXVIII, p. 266. As I have never since succeeded in obtaining a similar pre- 
paration I have concluded that in that case the granules were not of the normal composition. 
