September 5, 19 12] 



NATURE 



OUR BOOKSHELF. 



The Inter-Relationships of the Bryophyta. By 

 Dr. Frank Cavers. Reprinted from the New 

 Phytologist. Pp. vi + 203. Cambridge : At the 

 Botany School, 1911. Price 45.; postage 4^. 

 We are a little late in announcing that Dr. F. 

 Cavers's series of articles which appeared on 

 the inter-relationships of the Bryophyta in the 

 New Phytologist, vols. ix. and x., 1910-11, has 

 been issued separately. It is a great convenience 

 to have the work in this form, and it certainly 

 deserves this distinction. The classification is 

 mainly that adopted in Engler and Prantl's 

 " Naturlichen Pflanzenfamilien," but as a result 

 of his investigations the author introduces some 

 modifications. His proposed divisions are: (i) 

 Sphaerocarpales, (2) Marchantiales, (3) Junger- 

 manniales, (4) Anthocerotales, (5) Sphagnales, 

 (6) Andreseales, (7) Tetraphidales, (8) Poly- 

 trjchales, (9) Buxbaumiales, and (10) Eu-Bryales. 



Dr. Cavers discusses more particularly the 

 question of the old primary division of the Bryo- 

 phyta into two classes, Hepaticse and Musci, espe- 

 cially in relation to the Anthocerotales and the 

 Sphagnales. He argues : " If the Anthocerotales 

 are to be made a separate class apart from the 

 Hepaticae, either Sphagnales should also be con- 

 sidered a separate class apart from the Musci, 

 thus making four primary divisions of Bryophyta 

 — Hepaticae proper, Anthocerotes, Sphagna, and 

 Musci proper — or the Anthocerotales and Sphag- 

 nales might be united to form a class between the 

 Eu--Hepaticffi and the Eu-Musci, thus giving three 

 classes of Bryophyta." But he prefers dividing 

 the Bryophyta into ten groups as designated above. 

 The account of Riella capensis is of special 

 interest, and it is to be followed by a more detailed 

 paper on the' genus generally. Until 1902 this 

 singular aquatic genus was only known to inhabit 

 the Mediterranean region and the Lake of Geneva. 

 Since then a species has been discovered in the 

 Grand Canary ; another in Texas ; a third in 

 Turkestan; and a fourth in South Africa. 



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 cornniunications.] 



Determination of the Epicentre of an Earthquake. 



It has been proved by observation with the Galitzin 

 seismographs, both at Pulkowa and Eskdalemuir, 

 that when the first phase P of an earthauake is sharp, 

 the azimuth of the epicentre from the station is 

 uniquely determined by the observations at that station. 

 Itfollows that if the azimuth of the epicentre is deter- 

 mined at two independent stations suitably situated, 

 the epicentre can be determined from these two 

 azimuths alone. 



We have to-day, as an example, verified by con- 

 struction and by computation that this principle gives 

 accurately the epicentre of the earthquake that 

 occurred in Monastir on February i8, 191 1. 



The azimuth observed at Pulkowa was 22° 53' west 

 NO. 2236, VOL. go] 



of south, while the azimuth observed at Eskdalemuir 

 was 55° 56' east of south. The resulting epicentre 

 we find to be 40'5° N., 20'3° E. 



The epicentre deduced by the Pulkowa observations 

 of azimuth and epicentral distance was 4o'5° N., 

 20' 1° E. ; while the similar deduction from the Esk- 

 dalemuir results was 4o'3° N., 20'4° E. 



It is clear that in this case the accuracy of deter- 

 mination from the azimuths alone equals that of the 

 determinations from the separate stations, and 

 it is known that the earthquake did occur in the region 

 indicated. 



The advantages of this new method based on 

 azimuths alone are : — 



(i) That it is quite independent of any time reckon- 

 ing whatever at the two stations. 



(2) That it is independent of the determination of 

 the second phase S on a seismogram (which is fre- 

 quently difficult to fix with certainty). 



(3) That it is independent of any empirical tables 

 for epicentral distance, which are admittedly only 

 approximate. 



(4) Although only two stations are used, the deter- 

 mination is unique. 



We may observe that for a given case the accuracy 

 of determination depends on a suitable choice of the 

 two stations. 



B. Galitzin. 

 George W. Walker. 

 The Observatory, Eskdalemuir, Langholm, 

 Dumfriesshire, August 29. 



Implements of Man In the Chalky Boulder Clay. 



In Nature of August 15 Mr. Reid Moir has given 

 us certain interesting facts observed by him in con- 

 nection with the scratching of flints. 



(i) He notes the occasional scratching of what 

 remains of the "cortex" of the original nodule. It 

 does not seem to have occurred to him that such a 

 result may have been produced while the flint was 

 still enclosed in its original chalk matrix. Topley (in 

 his "Geology of the Weald") showed long ago that 

 the chalk strata had in many cases undergone con- 

 siderable difl'erential movement concomitant with 

 crustal movements ; and I have myself seen crushed 

 flints still in situ in the chalk cliffs at Ventnor, where 

 there is evidence of intense crustal movement of the 

 strata. So far back as 1880 I noted this, also the 

 extremely fractured and unworn condition of the 

 flints left as a residuum from the solution of the chalk 

 by carbonated rain-water on the top of St. Boniface 

 Downs (see P.G.A., vol. viii., No. 3), and my inter- 

 pretation of the phenomena there observable has since 

 been confirmed by Dr. A. Strahan, F.R.S., of the 

 Geological Survey. Here we have a sufficient 

 mechanical cause totally independent of anything 

 that may be connoted by the term " glaciation." 

 There seemed, moreover, to be just that slight 

 amount of surface-staining of the fractured surfaces 

 which might be due to meteoric iron-dust. 



(2) One fails to see that there is anv mystery about 

 the non-striated condition generally of the fracture- 

 surfaces of the flint fragments from the Boulder 

 Clay. How could the soft matrix of the Boulder 

 Clay scratch a flint, or even hold a harder stone with 

 sufficient grip to give it effect as a graving-tool, 

 however great the volume-pressure may have been ? 

 When the "glazier" wants to cut glass he does not 

 use putty to hold his "diamond." So much for the 

 talk of " intense glaciation " of hypothetical pre- 

 Crag flints, on which I hope to have shortly more to 

 say. 



On the other hand, boulders of the Chalk itseir, if 



