)54 



SCIENCE. 



[N. S. Vol. XV. No, 379 



are fully as slow as any of the negative 

 ones. 



At higher temperatures the air is ionized to 

 more than molecular distances from the wire. 



When the air was enclosed within a tube 

 the rate of discharge became very small. 

 Apparently particles are driven ofE from the 

 wire at the higher temperatures which are 

 suspended in the air within the tube. These 

 collect on the ions and greatly retard their 

 velocity. These particles do not aid in the 

 discharge, but materially diminish it. Their 

 presence may also be shown by their acting 

 as nuclei in the condensation of water vapor. 



These particles are found to be attracted 

 more by the negative ions drawn from a flame 

 than by the positive. It is, therefore, prob- 

 able that they cause the negative ions in the 

 discharge from the wires to have a smaller 

 velocity than the positive. 



Their presence is also shown when the wire 

 is heated in hydrogen, although to a smaller 

 extent. It therefore seems probable that they 

 are particles of platinum, and not of an oxide 

 of platinum. 



When the wire is first heated in a vacuum, 

 the discharge i^ much larger than at any time 

 afterwards. Heating the wire in hydrogen 

 largely restores to it the power of producing 

 discharge. At least some of the discharge 

 would, therefore appear to be caused by oc- 

 cluded hydrogen. 



The rate of discharge in a vacuum is much 

 larger than in air, but it was found to be 

 impracticable to find the velocity of the ions 

 in a vacuum. 



A complete account of the work will be 

 given soon in the Physical Review. 



C. D. Child. 



PALEONTOLOOWAL NOTES. 



NORTH AMERICAN ELEPHANTIDS. 



Any one who has had occasion to study 

 either the elephants or mastodons of North 

 America needs not to be told that the species 

 of each are very indefinitely known and, for 

 the most part, very imperfectly characterized. 

 Most of the species are based on teeth, one or 

 two on a single tooth, or at the best the 



description includes fragments of the jaw. 

 Specimens which have been gradually accu- 

 mulating in the U. S. National Museum make 

 it possible to at least commence the revision 

 of the species of our elephants, while the 

 material that has been gathered by the field 

 parties of the American Museum of Natural 

 History will throw much more light on the 

 subject. 



Of true elephants there appear to be three 

 good species, Elephas- primigenius, E. columhi 

 and E. imperator. The first-named, the north- 

 ern mammoth, a species of moderate size, 

 having teeth with narrow enamel bands, seems 

 to have ranged from Alaska southeasterly to 

 about the latitude of Washington, D. C. 



A line drawn from Washington to St. 

 Louis and thence northwestward to Victoria, 

 B. C, would roughly mark the southern 

 boundary of its habitat. To the south of this 

 line, extending to Florida and to the city of 

 Mexico, is found Elephas columhi, a much 

 larger animal on the average than the north- 

 ern species, having teeth with coarser enamel 

 bands. There seems to be an overlapping of 

 the two species, especially in the northwestern 

 United States, as noted by Professor Cope, 

 and along this line it is difficult at times, if 

 not impossible, to tell from which of the two 

 species individual teeth have come. Fully 

 grown examples of this species must have 

 attained a height of thirteen feet. 



Elephas imperator was based by Leidy on 

 an imperfect upper molar from the valley of 

 the Niobrara distinguished by its great size 

 and extreme coarseness of structure. This 

 specimen long remained unique and was final- 

 ly considered by Leidy to be the same as 

 E. americanus or, more correctly, E. columhi, 

 since the former name is unusable, being a 

 synonym. Last fall, however, Mr. W. H. 

 Holmes obtained in Indian Territory a con- 

 siderable number of teeth of both Elephas and 

 Mastodon from the same spot, comprising 

 molars of M. americanus, E. columhi and 

 some referable to Leidy's E. imperator. 

 Teeth of this species may be distinguished 

 from similar teeth of E. columhi by their 

 coarse structure, the large amount of cement 

 and the small number of enamel plates. Thus 



