302 Feecent Literature. [fas 
solve the problems to which Mr. Darwin first called attention.’ If such 
collections are not sufficient to throw light upon these problems, no col- 
lections will ever do so; and we cannot see how the discovery of five or 
six more subspecies of land-birds, or of some more accidental visitors, 
can alter our present conclusions.” These are: ‘‘I. The entire fauna of 
the Galapagos Islands [was] derived originally from America. II. It 
is uncertain whether there has ever been a iand connection between the 
various islands and between the islands and the continent or not.” In 
opposition, however, to Dr. Baur’s theory that the islands were once con- 
nected with America and with each other, and were submerged in or after 
the Eocene period, it is stated that the geological evidence ‘‘ is opposed to 
a former land connection with America” ; and that Dr. Baur’s supposition 
that the original number of species on this land-mass was small, and that 
as this mass of land became submerged, and the few original species which 
inhabited the whole area, having become restricted to the former mountain- 
tops, now islands, became differentiated in many different forms through 
isolation, is less reasonable than the hypothesis that they reached their 
present homes at different times from the neighboring mainland. 
Section V, ‘ The Birds of Galapagos Islands,’ occupies pages 142-199, 
and includes 108 species and subspecies, of which 9 are given as of 
either doubtful validity or doubtful occurrence. Of the 65 land-birds all 
but 5 are forms peculiar to the islands, and there are also nine water-birds 
peculiar to the islands. The remaining 31 species (excluding the doubt- 
ful forms) are for the most part wide-ranging seabirds or North American 
migrants. 
Fourteen new forms have been described from the Harris collection 
(mostly subspecies), of which six are described for the first time in the 
present memoir. The most noteworthy of these discoveries is the flight- 
less Cormorant (Pkhalacrocorax harrtst Rothsch.) from Narborough 
Island, where it was found only in the surf, its wings being too small to 
enable it to fly. 
Compared with Mr Ridgway’s ‘Birds of the Galapagos Archipel- 
ago’ (reviewed in this Journal, XIV, July, 1897, pp. 329, 330), there 
are numerous noteworthy differences, as with the 14 species added by 
Messrs. Rothschild and Hartert the total number recognized is only 
108, as against 105 in Mr. Ridgway’s list, while there are important 
differences in the nomenclature adopted. For example, Messrs. Roths- 
child and Hartert have for the first time used trinomials for the local 
forms of Passeres, explaining: ‘If trinomials are used everywhere else, 
there is no reason why the birds of the Galapagos Islands should be 
deprived of this most useful form of nomenclature. In cases where cer- 
tain individuals of representative forms are hardly, if at all, distinguish- 
able but where a series is easily separable, the recognition of subspecies 
is inevitable.” It thus follows that in many cases where Mr. Ridgway 
used a binomial, the present authors use a trinomial. Rothschild and 
