388 MAMMALIA FROM THE PURBECK OOLITE. [Ch. XX. 



It is worthy of notice, that in the Hastings Sands there are certain 

 layers of clay and sandstone in which numerous footprints of quadru- 

 peds have "been found by Mr. Beetles, and traced by him in the same 

 set of rocks through Sussex and the Isle of Wight. They appear to 

 belong to three or four species of reptiles, and no one of them to any 

 warm-blooded quadruped. They ought, therefore, to serve as a warn- 

 ing to us, when we fail in like manner to detect, mammalian foot- 

 prints in older rocks (such as the New Red Sandstone), to refrain 

 from inferring that quadrupeds, other than reptilian, did not exist or 

 preexist. 



But the most instructive lesson read to us by the Purbeck strata 

 consists in this : They are all, with the exception of a few interca- 

 lated brackish and marine layers, of freshwater origin ; they are 160 

 feet in thickness, have been well searched by skilful collectors, and by 

 the late Edward Forbes in particular, who studied them for months 

 consecutively. They have been numbered, and the contents of each 

 stratum recorded separately, by the officers of the Government Sur- 

 vey of Great Britain. They have been divided into three distinct 

 groups by Forbes, each characterized by the same genera of pul- 

 moniferous mollusca and cyprides, but these genera being represented 

 in each group by different species ; they have yielded insects of 

 many orders, and the fruits of several plants ; and lastly, they con- 

 tain " dirt beds," or old terrestrial surfaces and soils at different levels, 

 in some of which erect trunks and stumps of cycads and conifers, 

 with their roots still attached to them, are preserved. Yet when the 

 geologist inquires if any land-animals of a higher grade than reptiles 

 lived during any one of these three periods, the rocks are all silent, 

 save one thin layer a few inches in thickness ; and this single page of 

 the earth's history has suddenly revealed to us in a few weeks the 

 memorials of so many species of fossil mammalia, that they already 

 outnumber those of many a subdivision of the tertiary series, and far 

 surpass those of all the other secondary rocks put together ! 



Next anterior in age to the Purbeck mammalia are those of the 

 Lower Oolite at Stonesfield, to be mentioned at page 407. These are 

 all very small, comprising four species, three of which are certainly 

 marsupial, and the other possibly placental, but so unlike any living 

 type that some doubts are entertained as to whether it may not have 

 been marsupial. Still older than the above are some fossil quadru- 

 peds, also of small size, found in the Upper Trias of Stuttgardt, in 

 Germany, and more lately by Messrs. Charles Moore and W. Boyd 

 Dawkins, in beds of corresponding age in Somersetshire, which are 

 also of a very low grade, like the living Myrmecobius of Australia. 



If the three localities where the most ancient mammalia have been 

 found — Purbeck, Stonesfield, and Stuttgardt — had belonged all of 

 them to formations of the same age, we might well have imagined so 

 limited an area to have been peopled exclusively with pouched quad- 

 rupeds, just as Australia now is, while other parts of the globe were 



