( 410 ) 
The ray-florets, which are folded longitudinally in the bud in a 
peculiar manner, bear most water-glands on their veins, but for the 
rest the latter are distributed over the whole outer surface. These 
water-glands appear in much smaller numbers on the disc-florets and 
are limited on the corolla-tube to the junction of the narrow and 
wide portions. It cannot be determined at which stage of the bud 
the water-secretion begins. In very young capitula the drops escape 
observation, or may not yet be present. They cannot be observed 
until the outer whorl of the involucre curves outwards. During 
flowering the secretion of water cannot, of course, be observed on 
account of evaporation. It is not until after fertilisation, when the 
capitulum again contracts, and the involucral bracts grow out and 
again approach one another, that the secretion once more becomes 
evident. 
Coreopsis tinctoria Nutt., Coreopsis lanceolata L. Cosmos hybridus 
Hort. J ) 
In the genus Coreopsis and in Cosmos hybridus Hort., as in 
Dahlia the young capitula are enclosed and covered by the inner 
whorl of involucral bracts which are here generally light-brown. 
If the capitulum be opened at this stage, the florets in Coreopsis 
and in Cosmos hybridus are found to be in a bath of fluid as 
in Dahlia. 
In Dahlia the involucral bracts are superposed in two rows, 
without coalescing, but in Coreopsis and this species of Cosmos, 
arranged in a single row, they close up more tightly with their 
edges. This closing makes the extrusion of a drop here much rarer 
than in Dahlia. 
After fertilisation the capitilum does not close again in these plants 
and a secretion ot‘ fluid during the ripening of the fruit does not 
occur. 
Further investigation shows that here also the corolla is covered 
with the same long triehomes composed of 15 cells, as in Dahlia, 
but in these plants the young ovaries are free from triehomes. It IS 
very remarkable, however, that in Coreopsis and in Cosmos hybridus 
these triehomes do not secrete water, but a fluid which when heated 
with Iehling’s solution, gives a very marked reaction for glucose; 
in this respect therefore the secretion agrees with that found by 
Shibata in the calyx of Tecoma grandi flora. 
I doubt somewhat whether this plant has been correctly included in the genus 
Cosmos, \t appears to me rather to be a Coreopsis, 
