582 



SCIENCE 



[N. S. Vol. XXV. No. 641 



many others. The speaker stated that he was 

 unable to give any satisfactory explanation as 

 to how these arid land plants could have 

 crossed the barrier of the Cascade Mountains. 



In discussing this point, Mr. M. B. Waits 

 suggested that seeds of these plants might 

 have been carried by birds or possibly Indians. 

 The speaker admitted that both of these ex- 

 planations were possible, although he did not 

 regard either of them as likely. Mr. Vernon 

 Bailey called attention to the distribution of 

 many mammals in the Columbia Basin which 

 virtually coincide with that of the plants. 



Dr. Leonard Stejneger presented the next 

 paper on ' The Celtic Horse in Norway.' The 

 Celtic horse (Equus cahallus celticus) was 

 briefly described by Professor Ewart in 1902 

 as a small pony from the Outer Hebrides, 

 northwestern Ireland, Shetland, Fferoes and 

 Iceland, and more fully characterized in 1904 

 in a paper on the multiple origin of the horses 

 in Scotland, in which he compares it with the 

 'Norse horse' (E. cahallus typicus), the 

 essential differences consisting in different 

 proportions of forehead and snout and in the 

 absence of callosities on the hind legs of the 

 Celtic pony. The latter character at one time 

 was supposed to be of generic value since 

 these callosities are absent in the asses and 

 zebras. 



The speaker, during 1905, was able to 

 verify his suspicions that the fjord-horse of 

 western Norway is identical with the Celtic 

 pony by examining a number of pure-bred 

 west Norwegian horses which had no hind 

 callosities and which in other respects also 

 agreed with Ewart's descriptioa. The appar- 

 ent discrepancy between the views of the 

 speaker and that of Professor Ewart relative 

 to the identity of the Norse horse was ex- 

 plained by the fact that there are two native 

 races in Norway, the fjord-horse which is 

 identical with the Celtic, and the valley-horse, 

 or eastern Norwegian horse, which is the one 

 to which Ewart has reference. 



The important conclusion to be drawn from 

 the identity of the Scotch and the west Nor- 

 way pony is that it probably came to the latter 

 country from Scotland simultaneously with and 

 possibly domesticated by the West Norwegian 



brachycephalic population. This invasion 

 may have occurred in late pleistocene or early 

 post-glacial times, at the same period as a 

 complex assemblage of plants and animals, 

 known as the ' Atlantic ' biota crossed from 

 Scotland to west Norway over a continuous 

 land bridge, a question more fully discussed 

 in a paper now going through the press. 



In the discussion following Dr. Stejneger, 

 replying to a question, stated that no skele- 

 ton or part of skeleton of the Celtic horse is 

 in any museum in this country, and that even 

 in European museums, except that of the 

 Landwirthschaftliehe Hochschule in Berlin, 

 but little effort has been made to collect series 

 of the domesticated animals. 



Dr. Gill suggested that anatomical char- 

 acters derived from the laryngeal apparatus 

 might furnish better gi-ound for generic dis- 

 tinctions among the Equidse than the hind 

 callosities. 



M. C. Marsh, 

 Recording Secretary 



THE PHILOSOPHICAL SOCIETY OP WASHINGTON 



The 630th meeting was held on the evening 

 of March 2, 1907, President Hayford in the 

 chair. Mr. W. P. White read a paper on 

 ' Melting-point Determinations,' stating that 

 melting points are usually determined at 

 higher temperatures by heating the material 

 steadily and noting the stationary tempera- 

 ture which occurs during melting. With per- 

 fectly pure substances, this method presents 

 no difficulties or complications. With prac- 

 tically all substances, however, sufficient im- 

 purity is present to make the melting extend 

 through a temperature interval. This effect 

 is greater at high temperatures. The proper 

 curve is then distorted from two causes: (1) 

 The supply of heat from the furnace usually 

 varies. This trouble may be corrected by 

 regulating so as to keep the difference of tem- 

 perature between the furnace and charge con- 

 stant. The furnace is then held nearly sta- 

 tionary during the melting. (2) The interior 

 of the charge lags behind the outside during 

 melting; this effect becomes negligible if small 

 charges are used. An ordinary platinum 

 thermoelement can be put naked in a charge 



