The ermine was observed in the far north by early explorers and was mentioned in the literature, almost always under the name then used for the ermine of northern Europe and Asia. In 1896 Bangs misapplied to it the name richardsonii but Merriam in the same year corrected the application of this name and proposed as new for this weasel the name arctica, the name in use today. For almost 50 years after Merriam and Bangs wrote about it, arctica was treated, nominally at least, as a species distinct from its other relatives in both the Old- and New-World. The subspecific status of arctica was emphasized in 1944 (555) by the present writer in reporting in detail upon the specimens, of Mustela erminea, from Eastern Asia which were made available on loan by Professor B. S. Vinogradov and the late Anatol I. Argyropulo of the Leningrad Academy of Sciences. Specimens of Mustela erminea kaneii from the Asiatic side of Bering Strait and Mustela erminea arctica from the American side are distinguishable by slight cranial characters but in coloration and external measurements I can detect no differences. Merriam's (1896:16) mention of more golden-colored upper parts and darker underparts in American specimens than in erminea was the result of his comparison of Alaskan and northern European specimens. When Old World specimens from eastern Siberia, instead of from Europe, are used the differences mentioned by Merriam do not apply. Incidentally, many Siberian specimens have the white border, on the ear, which Merriam (loc. cit.) noted as a distinguishing feature of arctica. When Merriam named arctica he said (1896:15, 16) "Putorius arcticus . . . has heretofore been confounded with erminea or richardsonii. . . . It is interesting to find in this country an Arctic circumpolar weasel which, though specifically distinct, is strictly the American representative of the Old World erminea." Bearing in mind that Merriam's concept of species and subspecies (see Merriam, 1919:6) differed from that of nearly all modern systematists it is clear from his statement quoted above that he correctly understood the zoölogical relationship obtaining between the ermines of the Old and New Worlds.
Ognev (1935:31) seems to have been the first to use the name combination Mustela erminea arctica for Alaskan specimens. Thereby he expresses the view adopted here, namely that the American ermine is subspecifically but not specifically distinct from the Old World animal. Whether actual intergradation (crossbreeding) ever takes place across the narrow Bering Strait I do not know. I doubt that crossbreeding occurs but considering the Diomedes (islands), that might serve as a half way stopping point, and remembering Mr. Charles Brower's oral statement to me that he had seen tracks of ermine as far as 10 miles from the northern shore of Alaska out on the ice, the possibility must be granted of an occasional individual crossing from one side to the other of Bering Strait on the ice in winter or of being carried across when the ice broke up and drifted. If transfers of this kind occurred often one would expect ermines to occur also on Saint Lawrence Island where apparently they do not. The one skin (U. S. Nat. Mus. no. 259046) seen as labeled from there, my friend, Otto William Geist ascertained was imported as a skin with other furs from Siberia.
Ognev (op. cit.) who used the name combination Mustela erminea arctica for Alaskan specimens, applied it also to animals from Kamchatka. At the same time he recognized the animal from the eastern mainland of Siberia (as opposed to the peninsula of Kamchatka) under the name Mustela erminea orientalis Ognev 1928. Hall (1944:556) applied the earlier proposed name Putorius kaneii Baird 1857, to the animal on the eastern mainland of Asia and proposed the new name Mustela erminea digna for the ermine of Kamchatka. In comparing material of these two Asiatic races with topotypes and other specimens of M. e. arctica from Alaska, it seemed to me that the degree of relationship, one with the other, was about the same. M. e. digna has a slightly larger preorbital region than M. e. kaneii, and the skull is longer. In both of these particulars digna approaches closer to arctica. M. e. kaneii has longer tympanic bullae and a wider skull than digna and therein approaches more towards arctica than toward digna. As nearly as I can make out, digna and kaneii show a nearly equal degree of resemblance to arctica. Also the degree of difference between digna and kaneii is about the same as between either one of them and arctica. In view of the above considerations the ermines of the New and Old worlds are here regarded as only subspecifically distinct.
In the original description of Putorius audax (here regarded as inseparable from Putorius arcticus Merriam) Barrett-Hamilton erroneously designated the type locality as "Discovery Bay, North Greenland" whereas he should have written Grinnell Land [= Ellesmere Island of modern terminology] in place of Greenland. As reference to Nares (1877 and 1878) will readily reveal, Discovery Bay is near 65° W and 81° 40´ N, across Robeson Channel, to the west, from Greenland. The label on the type specimen and the specimen register in the British Museum of Natural History each designates the locality for this specimen, the type of audax, as Discovery Bay without mention of Greenland. The published accounts of Feilden (1878) and Nares (1877 and 1878) state that specimens of ermine were obtained at Discovery Bay. Probably H. C. Hart is the collector of the specimen; he was the naturalist attached to H. M. S. Discovery which wintered at Discovery Bay while H. W. Feilden was the naturalist attached to H. M. S. Alert which wintered a few miles southeast of Cape Sheridan, also on the eastern coast of Ellesmere Island.
It is true that from these ships a trip was made into Greenland and an ermine (only one individual it seems) was obtained there, but this individual was the type specimen of Mustela erminea polaris, in the account of which race something of the history of this specimen is given.
With the material available—and it is not entirely adequate—I can detect no features by which animals from the type locality of audax can be distinguished from typical arctica which latter name has priority.
Intergradation with richardsonii probably occurs completely across the continent. Intergrades here referred to arctica include those from Fort Goodhope. The one defective specimen from Lake Lebarge, Yukon, is not certainly identified as arctica and how far west of Teslin Lake the boundary-line between arctica and richardsonii should be drawn remains to be ascertained. The one specimen available from Hinchenbrook Island, no. 912 Mus. Vert. Zoöl., an adult female, is doubtfully referred to arctica because the damaged tympanic bullae appear to be no larger than in alascensis, and the size of the skull is more as in alascensis although intermediate between that race and arctica. Shape of the skull is more as in arctica. Possibly more nearly adequate material would show the existence on Hinchenbrook Island of an insular race differing in about the same degree from arctica of the mainland as does the insular kadiacensis. Nevertheless, the males from farther south at Cape Yakataga are in all respects arctica and this argues against near relationship to alascensis of the animal on Hinchenbrook Island. The three animals seen from Yakutat Bay are so young as not to display clearly the cranial characters of the subspecies but the extension of the color of the underparts onto the underside of the tail in them and also in the skin without corresponding skull from Glacier Bay, Alaska, is as in arctica, the race to which they are referred, and gives substantial basis for showing the geographic range of arctica as extending this far south along the Pacific Coast.