RELATIONSHII' OF ARMADILLO TO OTHER MAMMALS 223 



Of course Glyptodontoid and Armadilloid animals vary in their 

 bone-plating, as much as Rhinoceroses vary in their hide-plating ; so 

 do Horses and other mammals vary still more extensively in their 

 picture-plating. 



In some species of Rhinoceros the shields and plates have 

 entirely disappeared. Having got rid of the plates, some of the 

 species got rid also of the hide-shields. Natural selection was 

 evidently rendering their skin more pliable, in order that their 

 movements might be freer. In other words, their movements 

 became freer as they were relieved of their thick hide-plates and 

 shields. Not improbably the African species, as a compensation 

 for their loss of armour, may have obtained those formidable horns 

 which we see in the Natural History Museum. 



In the dappled Horse the pigment arrangements have been so 

 modified in most cases, as hardly to maintain any resemblance to 

 the typical rosetting of the Jaguar. In only a few cases have I 

 succeeded in tracing true rosettes in the Horse, and these I have 

 shown in Fig. 56. Usually it is hopeless to trace any minute rosette 

 resemblance in the Horse dappling. But by observing an infinite 

 number of all colours, and putting two and two together, one may 

 succeed in evolving an ideal of what some remote ancestor of the 

 Horse may have looked like. Anyhow, the Zebras are clear evi- 

 dence that, whatever the external disposition of pigments may 

 originally have been, the resulting modification has been almost 

 identical with that of the Tiger. In this, and much more clearly 

 in other Cats, it is easy to prove to one's-self that stripes and. 

 bands originated from rosettes like those of the Jaguar and 

 Leopard. 



Then in striped animals themselves we seem to witness modifi- 

 cations undergoing before our eyes, so to speak. In Grevy's Zebra 

 the bands are very numerous and narrow, some of them coalescing 



