336 BOTANICAL GAZETTE [NOVEMBER 
Erythea, these last at the base of the mountains. In the southern third of the 
peninsula many distinctly tropical genera appear, such as Ficus, Mimosa, 
Cassia, Albizzia, Jatropha, Haematoxylon, Lantana, Manihot, and Chiococca. 
ong the more remarkable endemic desert forms, two trees may be 
mentioned. One, alontteie to the Anacardiaceae, Pachycormus discolor, is 
found in the extremely arid central séction of the peninsula. Seldom ro ft. in 
height, the branches often shoot out twice that distance from the trunk, while 
their thickness (1 ft. or more), their abrupt ending in a few short twigs covered 
with red flowers, “reminding one of the proboscis of an elephant holding a 
nosegay,” give a remarkably grotesque appearance to the tree. The leaves 
are minute and fall off before the flowers are fully developed. The associated 
monotypic genus [dria columnaris is in the Fouqueriaceae, and appears as a 
tree reaching 50 ft. in height in a scattered open forest. In contrast with the 
preceding it bas a straight columnar trunk, usually without large branches. 
Illustrations of these and many other interesting and unusual plants add 
much to the interest of both reports—Gro. D. FULLER. 
Rubus in New England.—Brarnerp and PEITERSEN,” recognizing that 
“Rubus is one of the most polymorphic genera in the entire plant kingdom,’ 
have presented the blackberry group of that genus as displayed in New Eng- 
lan e authors say that the remarkable variation in the number of species 
Noa in oe various taxonomic works is due to too great reliance upon 
cimens, to failure to appreciate the variations due to environ- 
corial Sonica and to lack of appreciation of the extent of interbreeding. 
The present study is based upon data from material in the field, behavior in 
garden cultures and controlled plots, characters of the progeny of supposed 
natural hybrids, and behavior of progeny when artificially crossed. The 
result is that the authors recognize twelve valid species of New England black- 
berries, and a long list of hybrids: 
In following up the experimental work, PErrersEN*? has reached the 
following conclusions: variations ‘due to external factors are very marked; 
primordia of the prickle, glandular hair, and simple hair are present in all 
species; a large percentage of infertility occurs in most species, largely due to 
defective pollen; cross pollination is the rule in all species, all the species 
studied being either nearly or completely self-sterile; all the species are capable 
of inter-crossing under favorable conditions; duplicates of natural hybrids 
were produced en the progeny of a mucha of so-called species segre- 
gated as hybri 
The paper is a good illustration of the test of genetics applied to taxonomy. 
at. 
ae 
* BRAI INERD, Ezra, and Perrersen, A. K., Hgesoegian of New bcoz their 
classification. Bull. 217, Vermont Agric. Exper. Sta. pp. 84. pls. 36. 
7 PEITERSEN, A. K., Blackberries of New cena sent status oe the plants. 
Bull. 218, © aout t Agric. Exper. Sta. pp. 34. pls. 19. 
