NATURE 



49 



THURSDAY, MAY i8, 1905. 



THE BIRDS OF CENTRAL AMERICA. 

 Biologia Centrali- Americana. Aves. By Osbert 

 Salvin, M.A., F.R.S., and Frederick Ducane 

 Godman, D.C.L., F.R.S. 4 vols. (London: 

 1879-1904.) 



CONGRATULATIONS to the surviving author of 

 these volumes must be mingled with deep con- 

 dolence that his long-tried coadjutor and comrade 

 should not have been spared to complete this portion 

 of the great work in which they were jointly engaged, 

 and to supply that summary of its contents which 

 he, perhaps, alone could have written. But acutely 

 as the loss of Mr. Salvin is to be lamented, if on no 

 other account than this, no less real is the gratifi- 

 cation with which the bringing to an end of a task 

 that has lasted for a quarter of century is to be 

 regarded, and the relief to Mr. Godman 's mind at the 

 accomplishment of another portion of his gigantic 

 design must be enormous. It is getting on for twenty 

 years since the volume treating of the mammals of 

 Central .^Xmerica was reviewed in these pages by the 

 late Sir William Flower (Nature, xxxiv., p. 615, 

 October 28, 1886), and that portion also suffered by 

 the untimely death of its author, Mr. Edward R. 

 .\iston, so that instead of the comprehensive view of 

 the mammalian fauna of the country which he had 

 intended to appear in the introduction to the volume, 

 we had merely a series of tables of distribution which 

 he had prepared to found that view upon, and these 

 tables Mr. Sclater, who prefixed a few prefatory 

 sentences, left to speak for themselves. Speak for 

 tli'emselves they did, but they needed an interpreter, 

 since they were drawn up for the most part on 

 geographical lines — or, to be more accurate, from a 

 politico-geographical base, the geographical element 

 preponderating. 



The tables given in the first of the volumes treating 

 (if birds, and now before us (being almost identical in 

 form with those contained in the " Introduction " ' to 

 the first volume on Lepidoptera), do not differ very 

 greatly in character, though herein the political 

 divisions of the countrj' are given in greater 

 detail, so as to be more important than the geo- 

 graphical. Now each of these methods unquestion- 

 ably has its advantages — mostly of a practical kind. 

 If we want to see or obtain examples of any particular 

 kind of animal, it is convenient to know where it may 

 be found. But it can hardly be doubted that, had 

 .Mr. Salvin lived, he, with his experience of the 

 country and its ornithology, would scarcely have been 

 content without trying if it were not possible to treat 

 the distribution of the species, genera, and families 

 ;is well from a physical point of view. That he was 

 fully aware of the importance of taking that aspect 



^ That " Introduction " also contains a succinct description, excellent so far 

 as it goes, by Mr. Godman, of the natural features of each political district 

 of Central .America, which is taken to include the whole of Mexico from 

 the Rio Grande and the Rio Gila, but excluding Lower California, and 

 thence to the Isthmus of Darien in the now independent State of Panama. 

 The subject has been much more elaborately treated, though of course with 

 especial reference to the flora of the country, by Mr. Hemsley in his admir- 

 able ''.Appendix" to the fourth volume of the "Botany" of the whole 

 work (pp. 138-170). 



NO. T855, VOL. 72] 



of the question is shown by the pithy remarks on the 

 subject in an article published by Mr. Godman and 

 himself in The Ibis for 1889 (p. 242) — several years, 

 be it observed, after the appearance of Mr. Alston's 

 tables. The labour, no doubt, would have been 

 immense, and only to be performed by one possessed 

 of such knowledge, alike minute and wide, as Mr. 

 Salvin had; but assuredly he was convinced that it 

 can never be too strongly impressed upon all students 

 of topographical distribution that the key to the sub- 

 ject lies in the physical features of the country, 

 especially of a tropical country of such varied character 

 as Central America. Even an indication of the rough 

 division into the three well known zones — the tierra 

 caliente, the tierra templada, and the tierra fria would 

 be better than nothing, though in a country extending 

 over so many degrees of latitude and of such diverse 

 heights, what is the tierra templada of one district 

 becomes the tierra fria of aoiother. 



At the same time, it must be admitted that more 

 than this is required. Comparative altitudes and the 

 extent of forest-growths may explain some things, but 

 they will not account in all cases for the limits of the 

 area to which a certain form, say Pharomacrus or 

 Oreophasis, may be confined. But if boundaries are 

 not to be accounted for by physical characters, 

 assuredly they can be still less rationally explained 

 on political or geographical grounds. Considerations 

 of this kind seem to point to the futility of attempting- 

 to lay down any boundaries at all, unless those that 

 are physical can be traced, and of course the difficulty 

 of tracing them is sometimes very great. To take a 

 familiar instance here at home. Who can define on 

 physical grounds, or correlate with them, the distri- 

 bution of the nightingale in England and Wales? 

 Hence it may be fairly urged that it would be far 

 better for zoologists generally to leave off speaking of 

 areas, reg'ions, subregions, provinces, and the like, and 

 to regard the animal population of a country solely 

 from the faunal point of view. 



Central America would seem especially to lead to 

 some such conclusion as this. It can hardly be 

 doubted that the existing fauna of .America- -North 

 and South — is the result of at least three perfectly 

 distinct faunas, which have originated in, or been 

 derived from, as many different tracts, and probably 

 at as many different epochs. In Central .America 

 all three meet, though one is overwhelmingly out of 

 proportion to the other two. This is practically 

 identical with the fauna of by far the greater part 

 of South .\merica as distinguished from that of 

 Patagonia, which seems to have had a very different 

 origin and history, while the former is equally dis- 

 tinct from that which prevails now over the greater 

 part of North .America — this last being, as Prof. 

 Huxley long ago intimated (Proc. Zool. Soc, 1868, 

 p. JI4), much more closely allied to the Palaearctic 

 fauna, if, indeed, he might have added, it be not 

 substantially of Palaearctic extraction. Then again, 

 while comparatively few of the members of the fauna 

 now dominant in Central and the greater part of 

 South America have penetrated to the area at present 

 occupied by the apparently much more ancient Pata- 



T) 



