DISTRIBUTION OF SPECIES AMONG THE KEYS OF THE GROUP.
Only one other paper previous to the publication of the present one has dealt with the ecology of this region, viz, a series of maps with field notes collected by E. 0. Lansing in the spring of 1904, and published by Millspaugh in 1907.' This work, while fairly accurate and detailed. was compiled from notes evidently made in a very rapid survey of the islands, and in the four days (March 19 to 22 inclusive) allotted to the Tortugas group, the collector naturally overlooked many important minor features in the flora. For instance, evidence of haste is seen in overlooking the groups of sisal hemp, Agave sisalana (Engelmann) Perrin, among the bay cedars on Loggerhead, very old plants which were there even in the time of the third predecessor of the present light-house keeper, about 25 years ago; also on the same key numerous clumps and patches of Opuntia dillenii among the Suriana were not observed. Only one station for this plant is given for Loggerhead, viz, on the west coast near the light-house boat-shed, but the writer found it disseminated fairly well over the island in patches of old plants, often a meter high. These patches have certainly been there since the foun dation of the Laboratory of the Carnegie Institution of Washington on the key in 1904, according to statements of Dr. A. G. Mayer, the Director of the Laboratory, and the aged appearance of the plants supplements his statements. But in spite of these discrepancies, the paper has been of great interest to the writer, and, as Millspaugh says in a short introduction,' p. 191: "The principal value of the survey lies, therefore, first, in the historical record of the present flora, which should enable future students to determine which species have come to the different islands since 1904 and which have been unable to survive; second, in the knowledge of what species come first to such microcosms, thus forming a basis upon which to judge of the ease or difficulty of dispersion exhibited by certain species, and third, how and in what conformation species spread when brought into an untainted environment."
The paper has thus been of great value, to show what changes have taken place in the flora of the Tortugas in the twelve years which have elapsed since Mr. Lansing's survey—i. e., from 1904 to 1915.
The writer spent the two summers of 1915 and 1916 at the Laboratory in the Tortugas, taking field-notes, making some physiological experi ments on the plants, and working on the maps and collections of herbarium material. The herbarium specimens have all been de posited in the Herbarium of the University of Pennsylvania.
The scope of this paper has been confined to the Tortugas group because they are the most isolated of the Florida Keys and their flora is strikingly different from that of the Marquesas, 25 miles to the east, and also from that of the keys to the east of this latter atoll. The most conspicuous feature of this difference is the absence of the mangrove association to any considerable extent in the Tortugas, although Gar den Key has a few well-grown young trees which in the summer of the writer's residence in the islands produced flowers and fruits. Another reason was that while the writer had the opportunity and did visit and take notes on the keys above, as well as west of Key West, the longer periods spent in the Tortugas afforded much closer observation and facilitated the securing of experimental data. His work on the other keys was largely or entirely concerned with mangroves, some physi ological aspects of which have engaged his attention for several years.
The distribution of species on the various keys of the group is illus trated by maps made in the field by the author with the aid of a plane table and sighting-rule. The outline map was made in the survey of the group in 1914 and 1915 by Vaughan and The various keys will now be taken up separately and the species illustrated by symbols on the distribution maps. A comparison is made in each case with Lansing's survey.