DRAGONS OF OLD TIME 131 
powerful. Just as the birds and beasts (quadrupeds) of to-day 
show an almost endless variety, according to the circumstances 
in which they are placed, so that great and powerful order of 
reptiles we are now considering ran riot, and gave rise to a 
variety of forms, or types. 
Various attempts have been made to classify Dinosaurs and 
arrange them in family groups. But considering how imperfect 
our knowledge is of this very great and important order of 
reptiles, it is well to bear in mind that all such attempts are 
provisional and temporary. Much might be said in favour of a 
complete rearrangement of the classification usually adopted, 
which is that of the late Professor Marsh. As at present con- 
stituted, they appear to make a very artificial group. 
The order is usually divided into three sub-orders, as follows : 
(1) Theropoda, or beast-footed; (2) Sauropoda, or lizard-footed ; 
(3) Predentata, Ornithopoda, or Orthopoda. We shall follow 
this arrangement (used by von Zittel and others) but merely 
for convenience. Dr. H. Gadow raises them into a sub-class, 
making the above groups into orders, which perhaps is better. 
But it may be that Dinosaurs are not a homogeneous group. 
Perhaps some day the sub-order Sauropoda may be taken away 
altogether and placed with, or at least, near to the crocodiles, 
which they resemble in some ways. But Dinosaurs are strange 
beasts, very puzzling to the naturalist ; and, much as our know- 
ledge of them has been extended, future discoveries may bring to 
light forms quite as strange as those already known. Doubtless 
hundreds of specimens lie awaiting discovery in the unexplored 
fields of Asia, Africa, and South America. Dinosaurs also show 
obvious resemblances to the Ehyncocephalia and the Theromorpha. 
But nobody can say exactly where they should be placed. If we 
try to construct a genealogical tree of the reptiles we must put 
