1886.] BOTANICAL GAZETTE. 203 
still more restricted Prunus. That the physical conditions sur- 
rounding our island plants are extremely favorable for the rapid 
development, perfection, retrogression and perhaps ultimate ex- 
tinction of new species, certain observations of their habits tend 
to show. 
Of plants or species found conjointly upon the islands and 
mainland, the island forms are inclined to vary. The variation 
the most uniform and striking of all is in the preponderance of 
giant growths. 
rodizea capitata on Guadalupe, though restricted in area, was 
of such great size as to elicit surprise from Mr. Greene that it 
should have escaped the notice of his predecessor, Dr. Palmer. 
The same plant on Clemente, also within narrow limits, showed 
the same immense habit. Mr. Watson unqualifiedly referred it 
to B. capitata, only noting immense size and a trifling difference 
in the stamens, not enough to justify varietal rank. Its absence 
from Catalina, apparent absence from Guadalupe only ten years 
ago, and scarcity on Clemente, induces me to think that not only 
is it of recent introduction from the mainland, where in many 
localities its abundance gives character to the spring vegetation, 
ut that it is even now in a transitional state. That the pres- 
ence of identical physical conditions should elaborate similar 
forms on even widely sundered islands, is not improbable; hence, 
to quote the ideas, if not the words of the distinguished author 
of “ Plant Variations,” it is not difficult to believe, that on each 
island, within a few plant generations, we may witness the out- 
growth of a distinctively new type, sprung from a common stock, 
but different individuals, and varying from the parents with sim- 
ilar variations. This hypothesis would cover the case of the 
Prunus (the genus being continental), provided we could show 
co-existence at some past time of like conditions upon Catalina 
and its present West Indies habitat. : 
revert once more to the genus Lavatera as showing not only 
abnormal development of island species, but illustrating the facil- 
ity of some species to become exhausted or extinct when pa pably 
-uninfluenced by any other than strictly natural causes. : 
It is commonly known that very many plants, with skillful 
manipulation, “improve ” under cultivation ; ¢, ¢., at least increa 
the size of flower and leaf. This is anything but the case with 
Lavatera assurgentiflora, which I collected on Clemente from 
larger plants, in finer foliage and greater size and brilliancy of 
