June 26, 1913] 



NATURE 



The methods of qualitative and quantitative 



analysis employed by the chemist are not in- 

 cluded, since there is nothing specially applicable 

 to this particular industry in them. The historical 

 account of the apparatus devised for the purpose 

 of aerating- natural water by carbon dioxide is an 

 instructive example of the slow and gradual stages 

 required to effect a comparatively simple process. 

 From the first experiment of Bergmann in 1770 the 

 invention and elaboration of apparatus for aerating 

 and bottling has extended, and the latter half 

 of the book is devoted to descriptions of the 

 machines employed to-day. The commercial pro- 

 duction of liquid carbon dioxide has simplified the 

 process, and most ingenuity is centred upon the 

 charging and bottling machinery and the gas- 

 tighl fastenings. The description of the machinery 

 is somewhat superficial, and is the least effective 

 part of a commendable work. The examination 

 of mineral waters for bacteria and metallic con- 

 tamination is of special interest in view of the 

 various containers for such waters on the market. 

 Vorlesungen iiber allgemeine Histologic. By Prof. 

 Alexander Gurwitsch. Pp. v + 345. (Jena: 

 Gustav Fischer, 1913.) Price 11 marks. 

 Prof. Gurwitsch's work is not a text-book of 

 histology in the ordinary sense. It is arranged 

 in the form of a series of lectures, in which the 

 subject is dealt with not as an end, but as a 

 means to the solution of the wider problems of 

 biology ; a good deal of space, for instance, is 

 taken up with a discussion of the meaning of 

 heredity. The lectures, written as thev are from 

 a critical and philosophical point of view, are 

 full of interest, and examples are taken from every 

 branch of the kingdom of life to illustrate the 

 subject. The drawings of microscopic appear- 

 ances which beautify the text are numerous, 

 well selected, and well executed. W. D. H. 



LETTERS TO THE EDITOR. 

 [The Editor does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 can he undertake to return, or to correspond with 

 the writers of, rejected manuscripts intended for 

 this or any other part of Nature. No notice is 

 taken of anonymous communications.] 



Submerged Valleys and Barrier Reefs. 



In a letter on "Dana's Proof of Darwin's Theor) 

 of Coral Reefs," published in Nature for April 3, 

 Mr. Cyril Crossland points out that "land valleys 

 which extend beneath the sea arc not always proof 

 of subsidence. Such valleys, like coral reefs, may 

 owe their existence to different factors in different 

 cases." He adds that certain harbours on the east 

 coast of Africa "are the high parts of submarine 

 fault valleys," which, although thev simulate embay- 

 ments produced by the general depression and partial 

 submergence of a dissected coastal region, really 

 result from the local depression of fault blocks, and 

 hence cannot be taken as evidence that any coral 

 reefs which may occur near them have been built up 

 durintr a period of submergence. 



There can be no two opinions on this point ; but the 



discussion of fault-block depressions is aside from the 



problem involved in Dana's proof of Darwin's theorv, 



which is concerned with vallevs of erosion. To imply 



NO. 22/8. VOL. 91] 



that an argument which involves only such valleys 

 is vitiated because some "land valleys" are due to 

 down-faulting or other causes is unwarranted. If 

 "land valleys" due to faulting were called troughs" 

 instead of "valleys," their irrelevance would be more 

 apparent. 



Some of the harbours on the Red Sea coast, men- 

 tioned by Mr. Crossland as due to down-faulting, are 

 regarded by Mr. John Ball as partly submerged 

 valleys of normal erosional origin above sea-level; 

 so he states in a letter which appeared in Nature 

 lor Mav 22. Which of these two opinions is the 

 correct one may be best left to observers on the 

 ground, as neither of the writers here cited adduces 

 detailed evidence to support his conclusion. Mr. 

 Crossland's remarks on the relation of certain coral 

 reefs to abrasion 1 will not discuss here, because, 

 probably on account of the brevity of his note and the 

 lack of explanatory diagrams, his meaning is not 

 clear to me. But his statement that "land valleys 

 which extend beneath the sea are not always proof 

 ol subsidence" calls for comment, because it indicates 

 a misunderstanding of the- question at issue. 



The embayments considered in my article on 

 "Dana's Proof of Darwin's Theory of Coral Reefs," 

 published in Nature for February 0, were not such 

 .is occupy down-faulted troughs, or over-deepened 

 fiords of glacial origin, neither of which indicate sub- 

 sidence of their region, but only such as occupy valleys 

 of normal erosion ; that is, valleys which have been 

 excavated by the ordinary processes of subaerial 

 weathering and washing, and can therefore 

 have originated only on land above sea-level. The 

 peculiar and essential consequence of Darwin's theory, 

 which remained unnoticed by its author, is the 

 invasion of the previously eroded normal valleys of 

 a dissected and subsiding' coast by the sea ; and the 

 whole point of the long-neglected confirmation of Dar- 

 win's theorv lies in the evidence that Dana gave to 

 the effect that the drowned valleys of the Pacific 

 islands had been formed by the action of land waters 

 above sea-level before they were drowned, and not 

 by any other process, such as down-faulting or glacial 

 erosion, or by marine erosion, as Darwin apparently 

 thought. Hence, interesting as are the complications 

 which Mr. Crossland mentions, they do not touch the 

 question in discussion, which has to do, not with 

 shore-line embayments of whatever origin, but with 

 embayments of a highly specialised kind, occupying 

 valleys of normal erosion. So far as the evidence of 

 the Admiralty maps and of various recent observers 

 goes, the embayments of the central islands enclosed 

 by barrier reefs in the Pacific are practically all of 

 tli is highly specialised kind'; the occurrence of other 

 kinds of embayments elsewhere is no more relevant to 

 the case than the occurrence of upraised platforms of 

 marine erosion. 



The real point raised by Mr. Crossland's letter is the 

 possibility of distinguishing between embayments of 

 different origins. I cannot accept his opinion that 

 embayments'which occupy troughs produced by locally 

 down-faulted blocks simulate embayments formed by 

 the submergence of normally eroded valleys, unless in 

 a very rough manner, from which no confusion should 

 arise.' Even if the two kinds of embayments do in 

 some- superficial manner simulate each other, they can 

 he distinguished readily enough. A normally eroded 

 main valley is joined by branch ravines and side 

 valleys, all'systematically' related as parts of a valley 

 system; thev may have young, mature, or old forms, 

 according to their stage of development. Submerg- 

 ence of a main valley must therefore produce an in- 

 dented or branching embayment, like the " rias " of 

 north-western Spain or the drowned rivers of Devon- 

 shire-Cornwall. A down-faulted trough must at first 



