472 



NATURE 



[September 29, 1923 



I It 



.1 tr 



the in- 

 ns \»-=> 

 111*3/2, 

 length. 



1 ,0 (1 1 11. 'I 1(1 /) \i-. \\ !. 



Ill V el 1 hi- ll'.'ilt i!i;,i':' 

 ■ .111.1 A\ 



: 111. 1, 'I,.-!!. I. 



f if 5 be ^o 



ij^ht quaiituii; , close by Hk- -^un ut- 



ha\f ;>( 2 . io**gm., ^- 1-75 « ; ' ' hence 



wilh A/c= 2' iS-io'^'wefind AXas \ > hange 



wliirf ■ '— ton small to be f1' * nient. 



If. ur assume 11 lectron 



I'chr .1 \ (•r\' iniimtp , i .> V i , .. ii,i , ,. ,. though 



ai ' )■' .1 ' I ■!) Iiiil; t<> a diltcri'iit la\s on .n i onnt of its 

 clii' ' ... i>l't,mi .1 I I iiiucxiuu with A. I i . ( Oiiipton's 

 re. .i:! 'I ..i'i!ii[ii ;[;... i\- of thc scattfi'ini; of X ravs 

 (/'• Ma> 1923, n. .i.SO. .Milioiii'li this 



II lit form I' I I oiiiii t. ,:■ 1 he 



c\ ,, It is very li m cxpLnnuig 



the siiuili scaLLehug of 7-rays as ut 11 as their soften- 

 ing. To this extent it supports ihi- j)resent view 

 of the deflexion of light. I have worked out the 

 orbits of light corpuscles for a gravitating electric 

 charge, but the full discussion of the results and of 

 their bearing on the structure of the elect run imisL 

 be reserved for the paper already referred to. 



G. A. SCHOTT. 



University College of Wales, Aberystwyth, 

 September 3. 



Suggested Botanical Exploration of the Higher 

 Summits of the Cape Verd Islands. 



It is somewhat surprising that in our much ex- 

 plored world there is still a group of large islands in 

 the Atlantic which in a botanical and probably also 

 a zoological sense may be said to be imperfectly 

 known. For until we know what lives on the cloud- 

 capped summits of islands like Fogo and San Antonio, 

 which attain elevations of 8000 and 9000 feet above 

 the sea, it can scarcely be said that the Cape Verd 

 Islands have been scientifically explored. Surely 

 here would be a good piece of work for an English 

 yachtsman and two or three investigators from 

 Lisbon. Allowing two weeks for each island, the 

 examination of Fogo and San Antonio would only 

 involve about a month's absence from St. Vincent, 

 the assumed starting-place, and a host of botanical 

 and other curiosities would be gathered in the deep 

 ravines and on the uplands of those mountainous 

 islands. With government aid it could be carried 

 out by one of the learned societies of Lisbon. 



Fogo is reckoned to be the healthiest of the islands, 

 and it promises to be the most interesting for the 

 naturalist. But whoever goes will have to be pre- 

 pared for living in damp conditions, as in perpetual 

 fog and mist. 



Anything may be in hiding on those cloud-capped 

 uplands. Concealed in the ravines may still survive 

 plants that have become extinct in other Atlantic 

 groups, or which exist only in islands of other oceans 

 or in distant parts of the world, such as the Tree- 

 Compositae of St. Helena, the Tree-Labiatae of Juan 

 Fernandez, the Tree-Lobelias of the mountains of 

 tropical Central Africa. 



Then, again, American genera, like Clethra, that have 

 died out in the Canary Islands, may still survive 

 in the Cape Verd group, and the same may be said 

 of numerous other plant-types that have died out 

 in other parts of the world, or are almost extinct there. 

 Within the rain-belt of these mountainous islands 

 may still linger remains of once predominant laurel 

 woods and their associated plants, such as are now 

 so characteristic of the Azores, Madeira, and the 

 Canaries. 



The most significant features of island-floras are 



NO. 2813, VOL. I 12] 



presfiit<'i in thfi- 

 and it i^ on tills 

 hopes of jmportaiii 

 tion of the sumnii 

 th<- (aiM 

 I lawaii ! 



t.mt regions, 

 ly based the 

 arti^iiig iruiii the examina- 

 Mic mountainous islands of 

 .\K Thus, Tree-Lobelias link 

 Ocean with the highlands of 

 Kilin !i tropical Africa; the islands of 



J3oui! I ; .1 iiritnis in the Indian Ocean possess 

 spec ; . that are scarcely distinguish- 



able on in the Hawaiian mountains. 



So aK<ii>' im .iJiiMiii. ■> of the endemic genera of Juan 

 Fernandez connect its flora with St. Helena, the 

 Canary Islands, and th'- ' i>.<i..-.. i.i ...-is. The 

 distant connexions ol tii of the 



Socf>tr:m flora are ecjuall) : v cover 



mil' i.;lobe and are fou!: ica, and 



.\iii. : 



.\s llookiT iir<j.-d i- r floras, 



i-^lands have frecjiici.' ries for 



plant-types that I 1 :■ •...■; 



tinents, and in tl.< .>!,.. . ■ ;:;< 



(ape \'erd Islands to iiarbour the kilh aiul •;: .: 

 nian\ plant types that have failed in the stni^.,i< :<•: 

 existence in distant [larts of the world. 



We cannot afford to let slip opportunitie-s of this 

 kind for increasing botanical knowledge. The island 

 of Fogo has probably already lost much of its original 

 flora through the agency of the woodcutter, and not 

 many trees remain. Barker- Webb, Hooker, Schmidt, 

 Kransc, \'oi^cl, Christ, Hemsley, and Coutinho have 

 done iiiiK h to elucidate the plant-history of the Ca|>e 

 \'crd Islands, but thf^ .>ioc« i„t..^,..f ;,•,.", f....,,^..< of 

 their flora may be not 



Red House, Fo\ 

 Angus 



wall, 



Polar Temperatures and Coal Measures. 



I THANK Mr. Bonacina for his sympathetic 

 comment in Nature of September 22, p. 436, on my 

 letter on " Polar Temperatures and Coal Measures" 

 and for the added clarity he has brought to this 

 subject. He mentions disagreement with me on one 

 minor issue only, and that relating to the south polar 

 regions. In that connexion I am glad of the op- 

 portunity to confess that my thinking on the subject 

 of polar coal measures has really been based almost 

 exclusively upon my knowledge of the Arctic. My 

 suggestion that similar conditions might explain 

 Antarctic coal was a sort of parenthetical remark 

 made without any special consideration of the 

 Antarctic problem. 



Mr. Bonacina says : " I do not, however, fully support 

 Dr. Stefansson in expecting that a lowland south 

 polar continent surrounded by an ice-chilled ocean 

 would be liable, at least so often, to the high summer 

 temperature of the Arctic lowlands." A reading of 

 Mr. Bonacina 's letter in comparison with mine will 

 show that this partial disagreement is apparent only, 

 due to \\\\ faulty expression. I did not mean to say 

 that if the postulated low Antarctic continent were 

 somewhat larger than the actual present continent, 

 high temperatures would be as frequent there 2s 

 they would be in the Arctic if the land masses of North 

 America and Asia were connected across the North 

 Pole by continuous low land. All I meant to say 

 was that such a hypothetical low southern continent 

 might have temperatures high enough for the develop- 

 ment of a coniferous forest. 



Mr. Bonacina gives the explanation which I have 

 supposed correct for the lowering of summer tem- 

 peratures in the northern Mackenzie valley by 

 almost continuous winds blowing from the north. 



