376 MR. G. F. SCOTT ELLIOT ON THE EFFECT OF 
The following observations give some direct proof that this is 
really the case. Those made by me consist partly of a series 
of measurements made in Madagascar, and partly of the results 
given by an experimental garden, which I have had made this 
summer at home. I found on my return that Sorauer (in a paper 
published in Wollny's * Forschungen") had anticipated my con- 
clusions, and his results, as well as such other more indirect 
observations as I have been able to bring to bear on the point, 
confirm my own measurements. The first series of observations 
were made on the leaf-blade of grasses. In every case two sets 
of plants were gathered: the first (form A throughout the 
experiments) were taken from shady places, or had been sheltered 
during their growth by bushes or other plants, or were more 
thickly grown. The second (form B) were taken from the most 
exposed and driest spots I could find. The first Table shows 
the measurements which I made on a species of Agrostis, which 
turned out to be new (A. Elliotii, Hack. in Journ. Linn. Soe., 
Bot. vol. xxix. p. 65). These were preliminary and chiefly for 
the purpose of seeing how far the different leaves on the 
same stem (counting from below upwards) varied in this 
respect. 
TABLE I. 
Leaf-proportion in Agrostis Elliotii, Hack. 
| Length of Breadth of | Leaf- 
| leaf. leaf. ratio. 
| er med 
Second leaf (from base), A ...... | 16:6 36 4611 
Second leaf (from base), B ...... | 975 -32 3:047 
| Third leaf (from base), Å. > | 16:9 55 3:073 
| Third leaf (from base), B......... | 9:8 ss | 29 
| Fourth leaf (from base), A ...... 139 63 | 2206 | 
| Fourth leaf (from base), B ...... dol sa | w 
{ 
They show that the relative length varies greatly, and that the 
lowest leaves, which are, of course, most protected from wind 
and sun, are the longest proportionately. In all the following 
