74 



KNOWLEDGE, 



[Apbil, 1908. 



Although this is by no means an easy task, so far as 

 vertei>rates are concerned, it is the object of the present 

 article. As regards the phylogeny of particular family 

 groups of animals, attention is chiefly concentrated on 

 mammals— the highest of all animals — but the subject 

 would manifestly be incomplete without some reference to 

 the evidence as to the mutual relationship of the primary 

 groups into Avhich the vertebrata are divided liy naturalists. 

 It is this portion of the subject which (daiins our first 

 attention. 



With regard to the evidence in favour of a genetic 

 connection between vertebrates and lower forms of animal 

 life commonly termed invertebrates, opinions at the 

 jiresent day are so divided and so little is known with 

 certainty, that any lengthened discussion on this aspect of 

 the subject would he unprofitable. It may be mentioned, 

 however, that in the far distant epoch when the Old Red 

 Sandstone of Scotland was being deposited there flourished 

 in the lakes and rivers of the northern hemisphere a group 

 of mail-clad, somewhat fish-like creatures of which the two 

 leading types are collectively known as cephalaspids and 

 pteraspids. A striking feature of these lowly creatures, 

 and the one from which the former group takes its name, 

 is the presence of a large shield covering the head-region, 

 and in some cases taking somewhat the form of a cheese- 

 cutter. Now this shield has a curious resemblance to the 

 carapace, or shell, of a remarkable little freshwater crus- 

 tacean known as Apug, which may be met with in ponds 

 in this country. This resemblance is regarded by some 

 palaeontologists — Professor Patten among the number — 

 as not due to accident (if we may use the expression), but 

 as indicative of a true genetic relationship between the 

 two groups. And the writer whose name is cited in the 

 last sentence would even go so far as to say that we are 

 justified in regarding vertebrates and arthropods (crus- 

 taceans, insects, etc.) as diverging branches of a common 

 ancestral stock closely related to the cephalaspids. 

 Whether this striking theory will be confinned, or other- 

 wise, time alone will show, but it serves to indicate that there 

 is at least some kind of palaeontological evidence in favour 

 of the origin of vertebrates from lower types of animal 

 life. Nor are others wanting, although space and other 

 considerations do not admit of their discussion on this 

 occasion. 



We pass on, then, to consider the evidence in favour of 

 the descent of mammals from lower types of the vertebrate 

 stock. 



Till comparatively recently it was supposed that 

 salamanders— not the living forms, but their presumed 

 ancestors, the labyrinthodonts of the Carboniferous and 

 Permian epochs— were the nearest relatives of mammals, 

 and, on the evolution hypothesis, the representatives of 

 the parent stock from which the latter group originated. 

 This idea was in great part due to the circumstance that 

 while in reptiles (as in birds) the skull is aiticulated to 

 the first joint of the backbone, or vertebral column, by a 

 single knob — technically termed the condyle— in mammals 

 and salamanders there are two such knobs or condyles. 

 The single reptilian condyle consists, however, of three 

 parts, a median and two lateral elements, as is well shown 

 in the back view of the skull of a turtle. Now it is 

 evident that by the loss of the median element, and the 

 extra development of the two lateral parts, such a tripartite 

 single condyle could quite easily be converted into the 

 double condyles characteristic of mammals ; and, as a 

 matter of fact, there are indications how such a trans- 

 formation has been actually brought about. 



The possibility, or indeed thevery existence of such a 

 transformation would not, however, be sufficient to demon- 

 strate the descent of mammals from reptiles. To prove 



this we have to rely upon the evidence afforded by a 

 remarkable group of extinct reptiles which flourished 

 during the early portion of the Secondary epoch of 

 geological history, and attained a special development in 

 South Africa, or at all events, have left a large series of 

 their remains in that country. One of the mo.st curious 

 things about these anomodonts (or theromorphs), as they 

 are called, is the similarity of their teeth to those of 

 carnivorous mammals ; the division into incisors, canines, 

 or tusks, and cheek-teeth, or grinders, being just as well 

 developed as in the latter. But it is not on this dental 

 evidence, strong as it is, that we have alone to depend in 

 affiliating mammals to anomodonts. In the first place 

 there are certain very significant resemblances between 

 the skulls of the two, more especially as regards the 

 structure of the region of the cheek-bone. Of even more 

 importance is the community of structure displayed by 

 the bony girdles supporting the limbs of the anomodonts 

 and the same portion of the skeleton in the living egg- 

 laying mammals of Australia. In each, for instance, the 

 haunch-bone consists of three precisely similar elements ; 

 and an equally remarkable resemblance is displayed by 

 the shoulder-blade and the bones connected with the same 

 in the two groups. It is not easy to make those un- 

 acquainted with anatomy thoroughly realise the importance 

 of these structural similarities ; but to the anatomist they 

 are full of significance, and proclaim with no uncertain 

 sound the existence of an intimate genetic relationship 

 between the ancient anomodont reptiles and the modern 

 monotreme mammals. 



Dr. H. Gadow, of Cambridge, sums up the case con- 

 cisely, as follows : — " Mammals." he writes, " are descen- 

 dants of reptiles, as surely as the latter have been evolved 

 from amphibia (salamanders and their kindred). This 

 does not mean that any of the living groups of reptiles 

 can claim this honour of ancestry, but it means that the 

 mammals have branched where the principal reptilian 

 groups meet, and that is a long way back." The 

 anomodonts, or theromorphs, he adds, and especially the 

 smaller representatives of the sub-group known as 

 theriodonts, alone .show us what these primitive — half 

 reptile, half mammal — creatures were like. 



But, it may be objected, the most mammal-like of these 

 generalised reptiles were furnished with a full series of 

 formidable teeth simulating those of the carnivora, 

 whereas one of the two existing representatives of the egg- 

 laying, or monotreme, mammals, namely the spiny ant- 

 eater, is toothless, while the other (the duckbill, or 

 platypus) has only a few teetli on the palate in early life, 

 and these of an altogether peculiar and aberrant type. 

 There can, however, be little doubt that some, at least, of 

 the ancestral mammals were provided with a full series of 

 differentiated teeth like those of the carnivorous anomo- 

 donts, although it is by no means certain that the modern 

 egg-laving mammals are derived from such a type. 

 Among the anomodonts, for instance, there occur certain 

 forms with teeth not very unlike those of the duckbill, 

 and it is thus quite possible that the latter animal and 

 the spiny ant-eater trace their descent to a sjiecial branch 

 of the anomodont reptiles. 



For the ancestors of mammals other than the modern 

 monotremes we may not improbably look to the small 

 forms of which a few jaws have been met with in the 

 Stonesfield Slate of Oxfordshire. That these early 

 mammals were the direct descendants of the anomodonts 

 there can be little hesitation in admitting ; and it is also 

 quite within the bounds of probability that they laid eggs. 

 Unfortunately, however, our knowledge of these Jurassic 

 mammals is of the most vague and unsatisfactory nature ; 

 this being mainly owing to the extreme rarity and frag- 



