May 11, 1883.] 



SCIENCE. 



403 



pher. In the present paper he divides the scliistnse 

 rocJcs into micaceous, hornhlendic, and granulitic; 

 and the microscopic cliaracters of each group are 

 given in detail. He endeavors to show the relation 

 of the schistose rocks to the adjacent argillites, with 

 which they have been supposed to he continuous. 

 In the argiUite he found a few fragments of the horn- 

 hlendic rock, together with some felspathic fragments, 

 which, he says, came from a metamorphic series. 

 At another locality he found a fault (?) between the 

 hornhlendic rock and the argiUite, at which the latter 

 had been greatly broken. He states that the horn- 

 hlendic rock here resembles a greenstone, but thinks 

 he found in it signs of foliation and bedding. From 

 this evidence he draws the conclusions that the argil- 

 lites are younger tlian the metamorphic rocks 'by an 

 enormous interval of time,' and that, while the former 

 are devonian or older, the latter are azoic (archean). 



Without objecting at all to his conclusions, one 

 may point out the requirements to prove them, which 

 he has failed to give. He has not proved the schis- 

 tose series to be sedimentary, but admits that part 

 may be eruptive, and that some of the series, at 

 least, may be formed from volcanic ash. Until the 

 series is proved to he sedimentary, the finding of sup- 

 posed fragments of it in the argillites is no proof of 

 difference in age; for eruptive materials ai-e always 

 apt to be embedded in the rocks forming at the local- 

 ity at the time of the eruption. Bonney has further 

 taken foliation as bedding, with which it may or may 

 not correspond, and assumes that a metamorphic is 

 synonymous witli a sedimentary rock, when in real- 

 ity eruptive, especially basic, rocks are more easily 

 metamorphosed than most sedimentary ones; and 

 the former make a large part of the so-called 'meta- 

 morphic rocks ' in many regions of crystalline schists. 

 Until Bonney gives evidence to prove that liis series 

 is sedimentary, his conclusions cannot be regarded 

 as estahlished merely because he considers the rocks 

 sedimentary. 



The serpentine rocks of the Lizard district had been 

 discussed in a pievious paper, but additional material 

 is given here. Bonney holds that the serpentine is 

 formed from the alteration of an eruptive peridotite. 

 That serpentine is formed by the direct conversion of 

 olivine rocks has been conclusively shown by the 

 work of numerous lithologists; and, in this particular 

 case, Bonney's microscopic observations bear out the 

 general conclusion. That the Lizard peridotite was 

 eruptive was shown by its forming dikes in the ad- 

 jacent rocks, by its distorting and displacing them, 

 and by its enclosing fragments of them. — (Quart, 

 journ. geol. soc, ISS:!, 1.) m. e. w. [827 



METEOROLOGY. 



Thermometer -shelters. — There have recently 

 been worked up and published the results of experi- 

 ments, on a large scale, which were undertaken in 

 1860, by the Eoyal society of London, for the purpose 

 of testing various thermometer-shelters. The ex- 

 -periments were made in a large open field at Strath- 

 field Turgiss. Ten varieties of shelters were tried, 

 eight of these heing open, and two (Stevenson's .and 

 the Kew pattern) closed. It was found that all the 

 open stands were subject to serious objections, as 

 they gave varying results in different weathers. On 

 the whole, the closed shelters were regarded as the 

 better; and Stevenson's was preferred to the Kew, 

 as the smaller and more easily handled. It is still 

 thought, that, in dull weather, and for hygrometrical 

 observations, this screen has not sufficient ventila- 

 tion for the most accurate results. All the screens 

 gave nearly uniform results for the mean tempera- 



ture. Experiments are now in progress for compar- 

 ing wooden with Wild's metallic shelters. — (Enf/lUk 

 quart, weath. rep., l^I'd.) H. A. H. [828 



Terrestrial radiation. — Professor Tyndall placed 

 a thermometer upon cotton-wool whicli lay on the 

 ground, and suspended another four feet above it. 

 On Nov. 11, 1882, at p.m., the readings were: wool, 

 26'' F. ; air, '36°. There was nearly a dead calm, — sky 

 clear, and stars shining. The observations were re- 

 peated on Dec. 10, when, at 8.20 a.m., wool read 12°, 

 and air 27"=, with a clear sky, and very light wind. 

 In both instances snow covered the ground. On 

 many other days readings were made, and several of 

 these with the sky perfectly clear, and with no visi- 

 ble impediment to terrestrial radiation ; yel not one- 

 fourth of the difference was observed that occurred 

 on Dec. 10. Prof. Tyndall seeks to explain these 

 results by the hypothesis, long since advanced by 

 him, that the invisible aqueous vapor of the atmos- 

 phere in the latter cases interposes an effectual bar- 

 rier to radiation, and hence the difference. It would 

 seem as though a few observations of the amount of 

 vapor would have assisted in estahlishing or over- 

 throwing this supposition. 



In a later numher Prof. Woeikof discusses these ob- 

 servations, and suggests that the snow had a marked 

 effect in reducing the temperature of the air just above 

 it. He thinks that aqueous vapor has only a slight 

 effect in checking radiation; not, however, in its 

 gaseous state, but when condensed in small ice-crys- 

 tals or water-droplets, even if, which is sometimes 

 the case, it is invisible to the eye. He also suggests, 

 that, in order to determine the real effect of aqueous 

 vapor in terrestrial radiation, observations should be 

 conducted in a climate, where, with a relatively great 

 tension of vapor, the relative humidity is so small 

 that there is no dew on clear nights, or, at least, it 

 appears very late. Three thermometers, on cotton- 

 wool, should he placed, one on the ground, and the 

 others at heights from ten to a hundred feet above. 

 If Prof. Tyndall's views be correct, the highest ther- 

 mometer should show the lowest reading, as the aque- 

 ous vapor would impede radiation least from that one. 

 He thinks there would be very little difference be- 

 tween the three thermometers. The matter is cer- 

 tainly worthy of careful experiment. — (Nature, Feb. 

 15, March 15.) H. A. ii. [829 



PHYSICAL GEOGRAPHY. 

 Effects of deforesting in the Alps. — P. De- 



montzey describes, in a very well illustrated article, 

 the injurious results following the cutting down of 

 forests in the French Alps; these heing chiefly the 

 washing of great quantities of detritus down from 

 the slopes, the rapid formation of guUeys and ravines, 

 especially in the softer formations, .and the inunda- 

 tion of good valley-land with sand and gravel. The 

 extension of the torrential cone of the RIoubourdoux 

 (Basses Alpes), where the mountain stream enters a 

 broad valley, and several deep ravines, formed since 

 1830, on the branches of the Bourget, are excellently 

 shown. The remedy adopted against further growth 

 of the guUeys is to build numerous small dams across 

 the side streams, and thus force the waters to drop 

 their sediments, and build up their channels, instead 

 of deepening them. Planting trees is to go on witli 

 tliis as fast as possible, to prevent the wearing of the 

 b.are hillsides. — (La Nature, 1882, 151, 183, 215.) 

 w. M. D. [830 



Glacial erosion and lakes. — Rev. A. Irving has 

 recently read two papers before the London geologi- 

 cal society, — On the mechanics of glaciers, with spe- 

 cial reference to their supposed power of excavation, 



