2o6 



NATURE 



[December 15, 1910 



the parents are so remotely related that they are separated 

 by seven or more " free generations," the chances are 

 (unless one of the parents is highly prepotent) that, not- 

 withstanding the great merits of the immediate ancestors 

 and the expectations of Mendelians, the offspring will 

 revert to mediocrity. Hence it is not enough that breeders 

 should " mate the best with the best, avoiding close 

 affinities " ; they must avoid crossing distinct strains even 

 when the members of one strain closely resemble those of 

 another. This implies that, in addition to knowing the 

 pedigree of their stock, breeders should know as much 

 as possible of the wild races from which modern varieties 

 and strains were originally derived. 



J. C. EW'ART. 



Arctic Plants from the Vallsy Gravels of the River Lea. 



I HAVE recently found a plant-bearing bed in the Low 

 Level River-Drift of the Lea valley at Ponder's End. It 

 is exposed in an excavation worked by the Great Eastern 

 Railway Company, and I am indebted to Mr. Horace 

 Wilmer, engineer to the company, for permission to carry 

 on my investigations. 



The plant-bearing bed is found at a depth of 14 to 18 

 feet below the surface. It is embedded in stratified gravel 

 and sand, which presents much evidence of tumultuous 

 accumulation. In immediate association with it are found 

 tusks, teeth, and bones of the Elephas primigenius, Rhino- 

 ceros antiqtiitatis (if we are no longer permitted to call 

 it tichorhinus !), and other Mammalia. 



The pit is situated on the present floor of the valley of 

 the Lea at a level of about 35 or 40 feet above the 

 Ordnance datum. 



In correlation with the archaeological stages, the plant- 

 bearing bed of Ponder's End is later than the Mousterien 

 epoch. In fact, it is separated from this epoch by such 

 a wide interval that it is in all probability post-Palaeolithic. 

 On tlie other hand, it is unquestionably pre-Neolithic, 

 although the interval in this case appears to be compara- 

 tively short. There is thus little doubt that it comes 

 within the period of the archaeological hiatus between the 

 Palaeolithic and the Neolithic ages. 



It is by far the most important plant-bearing bed that 

 has hitherto been found upon this horizon within the area 

 occupied by Palaeolithic man in this country. It occupies 

 a position not represented on the well-known sites of 

 Hoxne or Hitchin. The only bed, so far as I am aware, 

 which can be placed on the same horizon is that at the 

 Admiralty Buildings, Westminster. This, however, only 

 yielded two species of plants, one of them being the Arctic 

 form Betula nana. 



I am at present engaged upon working out the botanical 

 material from this bed — a laborious task, occupying a large 

 amount of time. I am submitting this to Mr. F. J. Lewis, 

 who has very kindly undertaken its identification. A 

 considerable amount of material has already been ex- 

 amined, and, so far, Mr. F. J. Lewis has succeeded in 

 identifying ten species of plants, with four others doubtful. 

 Three of this number, namely, Salix herbacea, Betula 

 nana, and Sihhaldia proctimbens, are distinctively Arctic, 

 while most, if not all, of the remainder have a high 

 northern range, although thev are not confined to those 

 regions. Mr. F. J. Lewis defines the assemblage as Late 

 Glacial. 



The researches of Mr. Clement Reid in beds associated 

 with the PalfEolithic deposits have shown that there have 

 been many oscillations of climate in the south of England 

 since the deposition of the Chalky Boulder Clay. With 

 the evidence of this new bed before us there can be no 

 doubt that the Palaeolithic age was closed by a partial 

 return to glacial conditions, succeeding an epoch, or 

 epochs, when temperate conditions prevailed. This con- 

 clusion is in agreement with the results of recent work 

 upon the mammalian fauna of the Pleistocene age. 



This is not the place to enter further into this dis- 

 cussion. Enough has been said to indicate the importance 

 of this bed in throwing further light upon the climatic 

 changes of the Pleistocene age. It certainly suggests that 

 the archaeological hiatus is to be directlv associated in the 

 south of England with a final return of glacial conditions 

 of climate. 



If this view be sound, as I believe that it is, it seems to 



be perfectly justifiable to define Palaeolithic man as inter- 

 glacial, even although the last glacial phase above indi- 

 cated could not, of course, compare in severity with those 

 which preceded it. 



S. Hazzledine Warren. 

 Sherwood, Loughton, Essex. 



A New Theory of the Descent of Man. 



It is probable that some readers may fail to appreciate 

 Prof. Klaatsch's "New Theory of the Descent of Man'* 

 at its proper worth owing to the technical terms and 

 obscure descriptions used in the account published in 

 Nature of November 24 (p. 118). The theory is simply 

 this. The Neanderthal man and the gorilla' have con- 

 tinuous supraorbital ridges and similar markings for the 

 insertion of muscles on their skeletons; the Aurignac man 

 (who could pass as a fairly high type of modern humanity) 

 has not a continuous supraorbital ridge, in which he pre- 

 sents a very superficial resemblance to the orang, and has 

 certain muscular impressions on his skeleton somewhat 

 similar to the orang's. 



On this basis, which must be admitted to be " flimsy " 

 in the extreme. Prof. Klaatsch builds his new theory and 

 supposes that the gorilla and Neanderthal man are co- 

 descendants of one branch, the orang and the Aurignac 

 man of another. If one were to apply the principles used 

 by Prof. Klaatsch to the canine in place of the human 

 world, then we should say that the rough-haired New- 

 foundland is a co-descendant of a rough-haired bear, while 

 the smooth-haired mastiff has arisen with the sleek 

 leopard. An explanation is thus given of the points in 

 which the Newfoundland and the bear, the mastiff and 

 the leopard, have in common ; but what of the hundred 

 characters which the Newfoundland and the mastiff 

 possess in common, and which separate them from the 

 bear and leopard? Prof. Klaatsch ascribes these to "con- 

 vergence phenomena." At least that is how he accounts 

 for the fact that the Neanderthal and the Aurignac m- - 

 have all the features common to humanity ; one arose ; 

 the gorilla and the other arose via the orang, but bo: 

 arrived at the same structural goal so alike that most of ( 

 us regard them as the same species. 



The theory, owing to the demand it makes on " con- 

 vergence phenomena," passes somewhat beyond the limits 

 of rational speculation. Prof. Klaatsch's theory has failed 

 to gain the support of his able colleagues in Germany, and 

 is not likely to receive serious consideration in this 

 country. A. Keith. 



Royal College of Surgeons, December 10. 



The Cocos-Keeling Atoll. 



In reply to Mr. Wood-Jones's letter (Nature, 

 December i), I would say that I still consider that his 

 arguments against Sir John Murrav's theory go in support 

 of it. 



Mr. Wood-Jones suggests the reason for the precipita- 

 tion of calcium carbonate, when it has once begun, going 

 on until the solution contains • less than the normal 

 quantity ; what I wished to emphasise was that precipita- 

 tion does not begin until more calcium carbonate than 1 

 is normally present first passes into solution, i.e. that • 

 crystals can be formed in the interstices of the massi\ 

 corals in the lagoons until some of the dead coral is di^- 

 solved. There is, therefore, proof of solution in tlv 

 lagoons of atolls. 



In an early discussion on the same subject Sir Joli 

 Murray pointed out that the processes of the solution 

 the carbonate of lime of dead shells and skeletons by s 

 water, and of its secretion by the living organisms, a' 

 going on side by side wherever there are life and growr' 

 death and decay. In some regions secretion is in excf'' 

 and there is a formation of calcareous deposits ; in oth<^' 

 solution is equal to secretion, as in the red clay ar-r: 

 of the ocean ; in others solution may be in excess < 

 secretion, as in the larger and more perfect coral lagoons. 



In small coral atolls the periphery is large relatively : 

 the size of •'he lagoon, and the secretion of lime and t! 

 formation of coral sand are greatly in excess of the soli 

 tion that takes place, hence the lagoon becomes filled u; 



NO. 2146, VOL. 85] 



