436 



SCIENCE. 



[Vol. II., No. 34. 



I trust that the record which will be shown iu tlie 

 census report of cotton production, now in press, will 

 form a convincing illustration of the legitimate uses 

 of soil-analysis. E. W. Hilgard. 



University of California, Sept. 1, 1883. 



Do humming-birds fly back-wards ? 



The Buke of Argyll, in his Reign of law (p. 145), 

 lays it down in italics, that ' no bird can ever fly back- 

 zuardx.' He mentions the humming-bird as appear- 

 ing to do so, but maintains, that, in reality, the bird 

 falls, rather than flies, when, forinstance> he comes out 

 of a tubular flower. But this morning, while watch- 

 ing the motions of a humming-bird (Trochilus colu- 

 bris), it occurred to me to test this dictum of the 

 duke ; and, unless my eyes were altogether at fault, 

 the bird did actually fly backwards. He was probing 

 one after another the blossoms of a Petunia-bed, and 

 more tlian once, when the flower happened to be low 

 down, lie plainly rose, rather than fell, as he backed 

 out of and away from it. I stood within a yard or 

 two of him, and do not believe that I was deceived. 



It may not he amiss to add that the Duke of Ar- 

 gyll's objections seem to be purely theoretical, since 

 the ' Reign of law ' was published in 1866, and it was 

 not till 1879 that the author came to America and 

 saw his first living humming-bird. 



Bradford Torret. 



Boston, Sept. 14, 1883. 



"Wright's ice-dam at Cincinnati 



I notice on p. 320 of Science, vol. ii. no. 31, an 

 inaccurate report of what I said at the Minneapolis 

 meeting, which does injustice both to Mr. Wright 

 and to myself, and which I would beg to have cor- 

 rected. 



The reporter makes me speak slightingly of Mr. 

 Wright's discovery of the ice-dam at Cincinnati, as 

 not suflBcing to explain our Pennsylvania terraces. 

 On the contrary, I expressed my admiration for the 

 ■discovery as furnishing precisely the explanation we 

 need for the local-drift terraces of the Monongahela, 

 and the rolled-northern-drift terraces of the lower 

 Alleghany, Beaver, and upper Ohio rivers. 



The reporter probably mixed this up with what I 

 said afterwards respecting the rolled-drift terraces of 

 ■eastern Pennsylvania, which only reach a height of 

 SOO' A. T., in Northumberland county, and require 

 some explanation, perhaps, qiiite unconnected with 

 that which Mr. Wright certainly furnishes in a most 

 satisfactory manner for the 800' to 1,100' A. T. ter- 

 races of the Ohio River basin. J. P. Leslet. 



Second geological survey of Pennsylvania, 

 Philadelphia, Sept. 15, 1883. 



Erratic pebbles in the Licking valley. 



While engaged in tracing the outcrop of ' Clinton 

 ore ' in eastern Kentucky, in the fall of 1882, 1 became 

 interested in the pebbles, which in certain localities, 

 and up to a certain height, were very abundant in the 

 surface-soil. 



Most abundant were rounded quartz pebbles, prob- 

 ably from the millstone grit. Somewhat less abun- 

 dant were fragments of chert, showing little or no 

 wear derived from the sub-carboniferous limestone. 

 ■Still less abundant, though by no means rare, were 

 some from the carboniferous, often containing char- 

 acteristic fossils. They were confined, so far as I 

 <:ould determine, to the valley of the Licking and its 

 larger tributaries. Vertically, they range from the 

 river-bottoms to the top of the table, formed by the 

 upper Silurian rocks, which borders on the Devonian 



escarpment ; so that these tables are quite uniformly 

 covered with the material. 



The distribution of the material is such as could 

 only have been made while the valley was tempora- 

 rily occupied by a lake. I was therefore led, though 

 with some hesitation, to suppose that the glacier must 

 have crossed the Ohio at Cincinnati, damming the 

 river. I was not at the time aware of the labors of 

 Mr. Wright in tracing the glacier across the Ohio. 



Having now the certainty that there was a dam at 

 the required point, I think I may have no hesitation 

 in saying, that, during a portion of the glacial period, 

 the valley of the Licking was occupied by a lake which 

 overflowed laterally, and whose bottom became littered 

 with materials brought from the mountains of eastern 

 Kentucky by floating ice. They are most abundant 

 where the ice may be supposed to have had freest ac- 

 cess. 



Terraces which might have been expected are want- 

 ing in the region in which my observations were made. 

 Possibly they may be found in other parts of the 

 valley, especially above; their absence in the region 

 in question being due to the fact that only small por- 

 tions of the region would have reached above the 

 lake-level, which, by their disintegration, could fur- 

 nish the material for terraces. 



The overflow was probably to southward, hut I could 

 not search for it. Could it be traced, the amount of 

 erosion might give some data for an estimate of 

 time. G. H. Sqdier. 



Trempealeau, Wis., Sept. 14, 1883. 



Depth of ice during the glacial age. 



In the issue of Science for Sept. 7, reporting my 

 paper at Minneapolis, I am made to say, that, during 

 the glacial period, the ice was indeed " 600 feet over 

 New England, and very likely of equal depth over 

 the area to the west." I said 6,000 feet over Now Eng- 

 land. The evidences of glaciation are distinct upon 

 the G-reen Mountains to a height of nearly 5,000 feet. 

 The lower summits of the White Mountains, like Car- 

 rigain (which is 4,300 feet above the sea), are covered 

 with transported bowlders; and there can be little 

 question that some found by Professor Charles Hitch- 

 cock, within a few hundred feet of the summit of 

 Mount Washington, were transported thither by gla- 

 cial agency. Such is the evidence for New England. 



For the region north of Pennsylvania and the Ohio 

 River, direct evidence of such a great depth of ice is 

 naturally wanting ; but, according to Ramsay, glacial 

 scratches are numerous upon the summit of Catskill 

 Mountains in New York, at an elevation of 2,850 feet 

 above the sea. In southern Ohio there are numerous 

 places where the ice, within a mile or two of its 

 farthest extension, surmounted elevations which are 

 about 500 feet higher than the plains to the north of 

 them. I see no reason why it should not have been 

 as deep over the bed of Lake Erie as over the region 

 to the north of the White Mountains, though there 

 are there no glaciometers like Mount Washington to 

 measure the height of the frozen mass. 



G. Frederick Wright. 



Oberlin, O., Sept. 13, 1883. 



The ' stony girdle ' of the earth. 

 In your issue of Sept. 7, just received, you are kind 

 enough to insert a synopsis of the two abstracts of pa- 

 pers which I sent to the Minneapolis meeting. Allow 

 me the space necessary to make a correction and some 

 brief explanations. We are required to furnish these 

 ' abstracts ' to suit a printed form of small note size, 

 which is apt to lead to small chirography: hence I 

 suppose the mistake in reading and printing the title. 



