THREE EXTINCT SPECIES OF ELEPHANT. 301 
not more than two annuli, which are of large size and with thick enamel, whilst in fig. 9 
the corresponding plates show five or six annuli, of small size, and a corresponding thin 
enamel. The above differences appear to me to be quite as important as those which 
exist between the true molars of the Indian and African species, and infinitely greater 
than those which distinguish the former species from E. primigenius. 
In fig. 10 is shown the crown surface of a tooth from Maccagnone, which would seem 
in its size and characters to approach very nearly to the type of fig. 9; and I presume 
that Dr. Falconer may have introduced the figure with the view of showing some relation 
between the Sicilian and Maltese teeth. He has, however, left no observations on this 
point ; and as the tooth itself is not now with the others, I am unable to say more about 
it than is shown in the figure. 
With regard to the milk-teeth, we have not the same facility of judging of their 
relations from the thickness of the plates alone as we have in many cases in the true 
molars. In this respect little or no difference will be observed among the various 
milk-teeth of which figures have been given; and some might thence, under the 
circumstances, be led to conclude that these teeth must all belong to one only of the two 
species whose molars differ so widely in the thickness of the plates. But such a con- 
clusion is by no means warranted by what we know of the milk-teeth in different species 
of Elephant, in which, notwithstanding very great differences in the thickness of the 
plates in the true molars, little or none will be found in that of the plates of the 
respective milk-teeth, whilst in some cases the difference in this regard will even be in 
an opposite direction to that in which it might be expected to showitself. For instance, 
the mean thickness of the true molar plates in EZ. africanus varies from 0-85 to 1” or 
more, and in £. indicus is about 0'-55, whilst the thickness of the plates in the m.-m. 3 
of the two is pretty nearly the same, or about 032-33; in EZ. primigenius, although 
the thickness of the plates in m. 1 is not more in most cases than about 0"-43—-45, thase 
of the 5rd m.-m. average about 0-34, or rather more than in E£, africanus, though con- 
siderably less than in . antiquus,in which the thickness may be taken at about 0"-41, 
As we cannot, therefore, rely solely upon the thickness of the plates in the milk-molars 
as a diagnostic character in species so widely distinguished as FZ. indicus and E. africanus, 
it is impossible from that character to say whether or not all the milk-molars in the 
Zebbug collection belong to one or more species. Had they all been so worm as to 
afford a good view of the form of the machzrides, the decision would probably have been 
easy enough ; but it will be observed (leaving out of the question the m.-m. Z, fig. 2) that 
only three of the specimens were worn sufficiently for this purpose; and as, from the 
form of the machzrides in these instances, and the general condition and colour &c. of 
the teeth, it is not improbable that all belonged to one and the same individual, we must 
have recourse to other characters to determine the question of the true relations of the 
unworn specimens. Iam not sure how far such a character may be depended upon ; 
but, in the case more especially of milk-teeth uncovered with cementum, I think what 
ar2 
