March 27, 1903.] 



SCIENCE. 



511 



Geranospizse (1S73).* CircinEe, part? (or 



Circaetinae, part ? ) . 



Urubitingoe (1873). Urubitinginoe.t + 



Buteoninse, part. 



Buteones (1873). Buteoninse.t part. 



Haliaeti§ (1873). Milvinae. 



Aquila>|| (1873). Aquilinae. 



CircaetiH (1873). Circaetinse. 



Archibuteones** (1873). Buteoninse, part. 



Morphni (1876). Thrasaetinse. 



The only group of Mr. Pycraft's classifica- 

 tion having no equivalent in my arrangement 

 is his subfamily Vulturinse (comprising the 

 genera GypoJiierax, Neophron, Gyps, Yultur 

 and Otogyps^\). This subfamily he locates 

 thougli the latter he places also in Circoetinae! 

 JJrotriorchis I have not been able to examine, but 

 Geranospizias is certainly not closely related to 

 Circus, but seems to come very near to Polyio- 

 roides. 



* My Geranospizse included Polyboroides, a 

 genus not mentioned by Pycrait. 



t Whether Pyeraft would include more than 

 Urxibitinga is uncertain. My group contained, in 

 addition to that genus, Buteogallus, Heterospizias, 

 and Paraiuteo ( ' Anterior ' ) , the last of which 

 Pyeraft places in his Buteoninae, the other two 

 not being mentioned by him. 



t Pycraft's Buteoninae includes Archihuteo, 

 which I had placed by itself, Parabuteo ( ' An- 

 tenor ' ) , which I placed in Urubitingae, and 

 Busarellus, which I put with Haliaeti. 



§ My Haliaeti included Thalassoaetus, Mali- 

 aeetus, PoUoaetus, Ealiastur, Milvus, and Busarel- 

 lus, to which I would now add Gypoictinia. 

 Pycraft's Milvinae includes Haliaeetus, PoUoaetus, 

 Ealiastur, and Milvus, to which are very doubt- 

 fully added lotinia and Bostrhamus. 



II My Aquilse at first included, besides the genera 

 comprising Pycraft's Aquilinae, Harpyhaliaetus, 

 Morphnus, and Thrasaetus, but in 1876 the last 

 two were taken out and designated as a separate 

 group, Morphni, exactly equivalent to Pycraft's 

 Thrasaetinae. The correct position of Harpy- 

 haliaetus is, with me, a matter of doubt, but I am 

 now inclined to the opinion that it should either 

 go into the Urubitingae or constitute a monotypic 

 group. 



U My Cireaeti consisted of Circaetus, Spilornis, 

 and Helotarsus ; Pycraft's of the first and laat, 

 the second not being mentioned by him. 



** Consisting of Archhuteo only. 



tt It would be interesting to know where Mr. 

 Pyeraft would place Gypaetus. 



between the Thrasaetinse and Circaetinse, a 

 position not far different from that I would 

 have given it had occasion required, as is in- 

 dicated on page 227 of my ' Outlines.' 



That Mr. Pyeraft was unable to give the 

 preparation of his paper the amount of time 

 and care which the subject would have justi- 

 fied is obvious from several slips, nomencla- 

 tural and otherwise. For example, he places 

 the Polybori (his Polyborinse) both in the 

 Falconidse and Buteonidse (p. 315), and Ger- 

 anospizias in both Circaetinse and Oircinsel 

 In different places the terms Accipitridse and 

 Buteonidse are used for the same family. 

 There are also some errors in the explana- 

 tions to the plates, fig. 10, pi. 82, represent- 

 ing Catharistes, not Serpentarius, Pig. 11 on 

 the same plate being the latter, though not 

 so indicated in the text on p. 320; while Fig. 

 5, pi. 32, is Polyhorus, not Ihycter, as stated. 



On the whole, Mr. Pycraft's paper is an 

 excellent and most important contribution 

 to a very interesting ornithological subject, 

 and it is to be hoped that after extending 

 his investigations to numerous forms not 

 mentioned by him and therefore presumably 

 not examined, he may finally give us the 

 benefit of his studies in a more elaborate 

 treatise. 



Egbert Eidgwat. 



U. S. National Museum, 

 December 11, 1902. 



ELEPHAS COLUMBI AND OTHER MAMMALS IN THE 

 SWAMPS OP WHITMAN COUNTY, AYASHINGTON. 



On the 27th of November, 1877, on my 

 way down the Columbia Eiver, from the 

 Dalles, Oregon, I met an army surgeon who 

 told me of a deposit of extinct animals, dis- 

 covered the year before in ' mud-springs,' in 

 the swamps of Pine Creek valley. Whitman 

 County, Washington, about 100 miles north 

 of Walla Walla. Mr. Copeland, in probing 

 one of these springs on his farm with a long 

 pole, thought the end entered the occipital 

 foramen of a large skull. He had a long iron 

 rod with grappling hooks at the end made. 

 With this tool, and with the assistance of his 

 neighbors he was able to dislodge and bring 

 to the surface a very complete skull of the 



