1888.] Prafulla Chandra Ray on Copper- Magnesium Croup. 281 
There is generally a considerable pause between the deposition of 
two successive crops, during which no crystallisation takes place, as I 
have already pointed out. 
Dr Gibson suggested to me that I should collect a given crop 
fractionally, and ascertain whether each fraction had the same com- 
position, and thereby obtain proof positive of the homogeneity of 
the whole crop. This suggestion I have attempted to carry out in 
preparations XVIII. and XX. The fractions deposited during each 
period of crystallisation were carefully collected, and in a number of 
cases weighed. These weights will be found under the details of 
the individual preparations in the early part of the paper. 
In XVIII. the numbers obtained by the analy&is of the first 
three fractions collected (a, and y) agree with those calculated from 
the formula Ee. 3Zn. 4[S04. (XHJ2SO4. 6H2O]. The analyses of 
the next four fractions (8, e, 4 ’?) gave results agreeing in each 
case with the formula 2Fe. 5Zn. 7[S04. (XH4)2S04. 6H2O]. Thus I 
have actually succeeded in collecting two crops of crystals having 
definite though different molecular compositions, in three and four 
fractions respectively. This behaviour has an evident and important 
bearing upon the question raised by Aston and Pickering, and 
cannot be explained it seems to me in accordance with their views. 
During the deposition of the first three fractions a considerable 
change in the composition of the mother-liquor necessarily occurred, 
and yet these three fractions proved on analysis to be all of the 
same composition. The numbers obtained by analysis agree closely, 
moreover, with those calculated for a simple compound of the two 
constituent double sulphates. After the deposition of the third frac- 
tion of this first crop, some time elapsed before any further crystal- 
lisation was observed. The results of the analysis of the four next 
fractions in which the second crop was collected, proved these four 
fractions to be all of practically identical composition, unmistakably 
different from that of the first three fractions, but agreeing closely 
with the calculated composition of the not very complex com- 
pound assumed above. The change in composition between the 
two “crops” was not in any sense continuous, as it should have 
been according to Aston and Pickering, but was distinctly abrupt, 
thus the percentages of iron found were 3’40, 3 -40, 3*42, 4*11, 
3*87 (?), 4*01, 4*19. 
