278 BOTANICAL GAZETTE [APRIL 
Acacia constricta. 
A series of experiments with this species was carried out, but 
it was found unfavorable for exact results, owing in part to the 
fact that its leaflets become tightly closed after wetting, thus rend- 
dering it difficult to secure perfect drying of the surface without 
overexposure and consequent uncertainty as to the true weight. 
Accordingly, the conviction that the data obtained were unreliable 
led to their rejection. For this second group, therefore, we are 
restricted to the positive results obtained from Parkinsonia and 
Prosopis, which exhibit either no capacity or very slight capacity 
for leaf absorption, so long as the leaves are in perfect condition 
and normally active. 
. GROUP III. 
This third group includes representatives of a number of genera 
much modified in form and structure, and differing among them- 
selves in their methods of meeting desert conditions. Several of 
these are more commonly seen without ‘than with leaves, photo- 
synthesis then taking place in their green shoots; while others, more 
dependent on leaf activity, are commonly in a leafless condition 
during a large part of the year, pushing out new leaves promptly 
when conditions are favorable, and dropping them again when they 
become adverse, .as is seen particularly in the case of Fouquieria. 
Holacantha Emoryi. 
Of this peculiar shrub a small branch with leaves was. cut and 
left several hours to dry. At the end of this time it was still fresh, 
with no indication of wilting. After weighing it was wet for two 
hours and thirty-nine minutes, after which it was weighed again, the 
weight remaining unchanged. Leaving the shoot now to dry until 
the next day, and then wetting it for four hours and twelve minutes, 
' there was a gain in weight of only o.5 percent. Part of the same 
shoot, destitute of leaves, was treated in the same way, and, after 
wetting four hours and eleven minutes, also showed a gain of barely 
©.5 per cent. of its former weight. These results indicate on the 
part of this species capacity for leaf absorption so inconsiderable 
that it may be neglected. 
Koerberlinia spinosa, a closely related species, agrees with Hola- 
‘cantha as far as observations have yet been made. Only leafless 
