November io, 1923] 



NA TURE 



681 



support from the arctic fauna, for while the land 

 mammals in polar regions are white in their snowy 

 environment, the pelagic mammals — whales, seals, 

 walrus, etc. — remain dark. The polar bear, hugest of 

 Ursidae, would encounter far more difficulty in stalking 

 seals — his favourite food — were it not for his white 

 mantle. 



Miss Pitt has undertaken useful analysis of the barn 

 owl's bill of fare. In twenty-eight pelts or castings 

 taken at random from the roosting-place of a bam owl, 

 she identified the remains of 112 small mammals and 

 3 small birds. " In less than a month that owl had 

 eaten 66 mice and rats and 46 shrews, a record that I 

 suspect few cats could equal." A cat, it may be noted, 



Fig. 3. — The liritish pine marten in full winter coat. 

 From ".Shetland Pir.-\tcs and other Wild Life Studies. 



might kill the shrews, but would not eat them, therein 

 showing a discrimination which it were well that 

 gardeners and others would observe between the 

 beneficent insectivore Sorex and the destructive 

 rodents Mus and Evotomus. 



Besides the experience gained through long hours of 

 vigil in a hiding-tent, Miss Pitt has made still more 

 intimate acquaintance with many wild animals, not as 

 mere pets, but as free companions and messmates. Of 

 these, the most intellectual were a pair of ravens, which 

 spent much of their time " ragging " the cook alter- 

 nately with her cat ; the most docile was a merlin 

 hawk, the most playful a pine marten (Fig. 3), which 

 came as a " kitten " from the Cumberland Fells, and 

 fjuite the most foolish and awkward was a brown 

 hare. There is much entertainment, as well as sound 

 information, in both these volumes. 



Herbert Maxwell. 



NO. 2819, VOL. I 12] 



Earth and Sun. 



Earth and Sun : an Hypothesis of Weather and Sunspots. 

 By Ellsworth Huntington. With a Chapter by 

 H. Helm Clayton. Pp. xxv + 296. (New Haven: 

 Yale University Press ; London : Oxford Univer- 

 sity Press, 1923.) 235. net. 



FOR half a century or more, it has been known 

 that the earth's magnetic condition varies]^in 

 striking similarity with the state of activity on the 

 sun's surface. Many attempts have been made to 

 establish similar connexions between meteorological 

 phenomena and the sunspot cycle, but only within 

 recent years has it been possible to record indisputable 

 success in such attempts. The ele- 

 ment most clearly affected is, as 

 might have been expected, the tem- 

 perature. Koppen's work, sup- 

 ported by that of several other 

 writers, demonstrates that at sun- 

 spot maximum the mean tempera- 

 ture of the atmosphere is slightly 

 less than at sunspot minimum. The 

 difference is small, being o°-6 C. in 

 the tropics, and falling to o°'4 C. in 

 temperate latitudes. It seems not 

 unlikely that the diminution at sun- 

 s])Mt maximum corresponds rather 

 to increased terrestrial absorption — 

 due to a greater amount of ozone 

 in the upper atmosphere — than to 

 diminished output of radiation from 

 the sun. The sun sends out in- 

 creased corpuscular emission, and 

 almost certainly increased ultra- 

 violet radiation, at times of sunspot 

 maximum, so that it would be rather surprising were 

 its total radiation to be diminished at such times. 

 On the other hand, intensified short-wave radiation 

 would probably produce more ozone, which would 

 intercept a larger proportion of radiation on its way to 

 the earth's surface. 



Small as is this temperature variation, it may be 

 expected to produce important effects upon other 

 terrestrial phenomena. Such effects would show a 

 connexion with the sunspot cycle, possibly almost 

 as close as that siiown by the temperature varia- 

 tion itself. Hence the fact that a meteorological 

 phenomenon is strongly correlated with the solar 

 activity does not necessarily imply that the connexion 

 is direct and independent. It is doubtful whether any 

 other independent solar meteorological effect has yet 

 been established, though some remarkable secondary 

 effects are known. For example, Mr. C. E. P. Brooks 



T I 



