330 
the part of For, but does Dr. HEIDENHAIN in this case attach no 
weight to the cumulative negative evidence against it? 
It is perfectly true, in plants, that centrosomes have been descri- 
bed by one author for Lilium, but all who have sought for them 
since have never been able to find them. Dr. HEIDENHAIN, in criti- 
cising my own work, suggests that I ought to have tried control experi- 
ments. I may be permitted to remark that it is precisely as the 
result of control observations that I have been led to adopt the view 
which he quotes (p. 241) respecting this plant. I have seen and 
described centrosomes in a number of plants in which they certainly 
occur, and I have also investigated a number of animals on purpose 
to become familiar with these bodies. Dr. HEIDENHAIN hints that my 
failure to discover them in other cases may be due to imperfect or 
improper fixation or to inadequate staining. I may however explain 
that I never depend merely on the results obtained from one or two 
fixatives, and I am perfectly aware of the danger of relying on any 
single method of staining. I am quite cognizant of the errors into which 
an unskillful use of iron-haematoxylin may lead one, and I can assure 
Dr. HEIDENHAIN that, however highly I may appreciate them, I have 
never trusted to his own methods of staining alone, but have con- 
firmed and checked my results by a comparison of the effects of other 
reagents and stains. 
Indeed I would go so far as to recommend Dr. HEIDENHAIN to 
study the pollen-mother cells of Lilium for himself and then his 
criticisms will have the merit of being based on direct observation 
and not on gratuitous assumptions. 
For my own part, I have not ventured to formulate my own views 
on these matters without devoting a considerable time to the study 
of both animal and vegetable cells, and if I have arrived at the con- 
clusion that it is at present premature to generalize widely on the 
presence or absence of centrosomes it is just because I find it im- 
possible at present to range all my observations on one side or the 
other. 
I may be permitted to add that Dr. HEIDENHAIN has obviously 
misunderstood the meaning of a passage which he cites on p. 241 in 
a foot-note. If he will refer to the original, he will find that the 
term “individuality” is applied to the centrosome in the sense of a 
spontaneously active, or automatic, entity, and hence his 
concluding criticism is devoid of meaning — indeed it rather damages 
his own case. My point was that although the centrosomes at 
