39<5 
Notes. 
Immediately following this stage in the nuclear division, the chromo- 
somes congregate in the equatorial plane, and simultaneously the 
granules already referred to in the earlier part of this communication 
are discoverable in the surrounding cytoplasm. They stain precisely 
as do the granules which arise in the nucleus by fragmentation of the 
nucleolus, and are very distinctly seen, many of them influencing the 
direction of the spindle-fibres as already described. It would at 
present be premature to attempt to do more than suggest that there 
may be a closer connexion between the granules of nucleolar origin 
and those which later occur in the cytoplasm, but it may be men- 
tioned that Hertwig 1 suggested the possibility of a nuclear origin for 
the animal centrosomes. There are certainly great difficulties in the 
way of accounting for the sudden and abundant formation of granules 
on the assumption of a purely cytoplasmic origin, and their obvious 
relation to the spindle-fibres, as well as their ultimate fate, opens out 
a whole set of further questions. I will not advert to them at greater 
length here ; I hope to be shortly able to speak more definitely on 
these, as well as other points raised in this preliminary note. 
J. BRETLAND FARMER, 
Royal College of Science, London. 
THE GENUS TREMAT O CARPUS 2 .— At Mr. Hemsleys request 
I have examined the structure of the fruit of Lobelia macrostachys. 
Dr. Zahlbruckner's note on his Trematocarpus macrostachys may be 
divided into two portions, the first of which is decisive for the second, 
viz. the structure of the fruit, particularly with regard to dissemination, 
and the generic value of the characters derived from that particular 
structure. The remarks which I have to make on this structure are 
based upon observations on the material preserved in the Kew 
Herbarium, and on a fruit sent by Dr. Zahlbruckner to Mr. Hemsley. 
The ovary wall of Lobelia macrostachys, Hook, et Arn., exhibits the 
same general structure as in allied species of Lobelia, the generic 
identity of which has never been doubted. One remarkable character 
is the presence of a well-developed system of vascular bundles ana- 
stomosing in a distinct network very like that found, for instance, in 
Z. nicotianaefolia , an Indian species, with the only difference that it is 
1 O. Hertwig, Die Zelle und die Gewebe, Jena, 1892, p. 48 and 164. 
2 See Annals of Botany, vol. vii. No. 26, and vol. vi. No. 21, 
