114 PLEISTOCENE MAMMALIA. 



the Lion ; but the variations in this respect in the skeletons of the animals which we 

 have examined in the British collections is so great that we cannot admit the specific 

 value of this distinction, and therefore we cannot assert that the average specimens of the 

 fossil humeri approach one species more than another. Singularly enough, however, the 

 two most perfect shafts we have seen differ considerably from those of any recent 

 Felis in their extreme antero-posterior compression, especially that from Oreston 

 (fig. I, c,d). 



The compression of the Gailenreuth specimen is not so evident. In the large fragment 

 ((?, c') w^e have figured from Bleadon, and in that of Dr. Schmerling (fig. 1, light tint), 

 there is no evidence of it whatever, and the many fragments we have met with show that 

 the general proportions of the shaft were those of the recent Lion or Tiger. Some are 

 intermediate between the compressed and the ordinary form, and connect the two 

 extremes together. In the absence, therefore, of any other evidence of a second large 

 species of Felis in the Pleistocene deposits of Britain and Germany, we refer the com- 

 pressed humeri to a somewhat abnormal form of the spelaean Lion. The comparison of a 

 large series of bones of the same species, either recent or fossil, shows that no particular 

 bone is cast in a crystalline form, but that the variations increase in proportion to the 

 number examined. How^ far this may go on without transgressing the limits of a species 

 depends obviously upon the judgment of the naturalist. If he believe that a species has 

 an actual existence in nature, he will look upon these variations as of specific value, 

 because they depart from the typical form, which he will take to be as invariable as the 

 figure of a crystal. If, on the other hand, he view a species as a mere arbitrary summing 

 up of points of agreement devised by man for the classification of the varied forms of 

 life, he will consider that the variations are simply the result of a more minute inquiry, 

 and he will extend the limits of his species to cover a very large amount of variation. 

 Few naturalists have recognised the amount of variation from a specific type, observable 

 in many individuals, on account of the immense labour required in the investigation. As 

 we hold the latter of these views, we consider that the compressed humerus is a mere 

 variation from the ordinary spelaean form. 



§ 3. Description. — The lateral aspect of the humerus is somewhat sigmoid, the 

 proximal end being bent slightly backwards, the distal slightly forwards. It is more or 

 less compressed proximally, and depressed distally, flat internally, and highly convex 

 externally ; the proximal articulation (fig. 1 a, a) is of considerable size, highly convex 

 posteriorly and internally, slightly concave close to the greater tuberosity {b). Its generally 

 convex outline is intercepted anteriorly close to the bicipital groove by a roughened 

 space devoid of synovial membrane, even with the surface of the articulation ; posteriorly 

 it overhangs the shaft, internally it is bounded by the lesser tuberosity of a reniform 

 mass, longer than broad, and extending from behind diagonally upwards to the level 

 of the highest part of the articulation, to which was attached the tendon of the sub- 



