470 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 48 
this horny callosity is transferred to the hybrids between Equus and 
Asinus. On this latter point I was able to make observations on 
hundreds of mules to the effect that the callosity is nearly always 
transferred to the progeny. This is important, since it renders it 
almost certain that in hybrids among the more closely allied species 
within the restricted genus—or subgenus—Eqvus, the callosity will 
always be found, if possessed by one of the parents. In these days 
of universal crossbreeding of the domestic races for their “ improve- 
ment” it has become very difficult to find a country horse of abso- 
lutely pure race except in the remotest part of mountain districts 
or regions otherwise difficult of access. 
Fortunately, Tscherski in his excellent work on some fossil Asiatic 
horses gave a clew by carefully describing both the external and 
the osteological characters of the “ Tarpan,”’ a small horse which 
until the middle of the last century was found wild in the steppes 
of south Russia. Not only did the tarpan lack the callosity on the 
hind legs, but in size and color also did it tally with the Celtic and 
the west Norway pony, and what is equally to the point, the skull 
“in its essential relative dimensions agrees with those of the Iceland 
horses.” 
The result of these preliminary investigations | summarized in a 
paper entitled “‘ Den celtiske pony, tarpanen og fjordhesten,” which 
was published in the June-July number of Naturen (Bergen), 1904, 
in which I urged the specific distinction of Equus celticus and sug- 
gested the extreme probability that all three horses mentioned, viz., 
the Celtic pony, the Russian tarpan, and the west Norway fjord 
horse belong to this species, and the various heavy European horses 
to another species which must stand as Equus frisius (Boddaert).? 
I also suggested that the fjord horse came to west Norway from 
Scotland as a descendant of the Celtic pony. 
About the same time Professor Ewart published a more detailed 
*Mém. Acad. Sci. St. Pétersbourg (7), xt, No. 1, 1892, pp. 257-383. 
*Of which Fitzinger’s Equus robustus and Nehring’s E. caballus var. 
germanica are synonyms. 
The Linnean Eqvus caballus is a collective name without indication of type 
specimen, of course. The restriction of the name, as practiced by the first 
revisor, must, therefore, be accepted, and this restriction it appears was 
undertaken by Fitzinger, in 1858, who reserved the name E. caballus for the 
group having the Arab for type. Now to restore the name to the horse occur- 
ring in eastern Sweden is clearly impracticable, as no well-defined form is re- 
stricted to that country and, moreover, would be contrary to all accepted 
nomenclatorial codes. The relationship of Fitzinger’s E. velox to his E. 
caballus is one of the many unsettled questions in European hippology. 
