June io, 1909] 



NA TURE 



429 



so far as possible, the nature of the earliest fishes which 

 appeared in Silurian times, and compare them with the 

 type of arthropod which had been evolved up to that 

 lime. Ammoccctes was chosen rather than Amphioxus 

 because it resembles the extinct Cephalaspids more closely 

 than does any other living fish, while, on the other hand, 

 Limulus is the only living example of the great arthropod 

 group which dominated those Silurian seas, a group which 

 gave origin to both arachnids and crustaceans, and was, 

 of necessity, nearer to the ancestral annelid type than 

 most of the arthropods of the present day. In the attempt, 

 then, to generalise the characteristics of such a group, it 

 naturally follows that account should be taken of the 

 structure of annelids and of such a low type of arthropod 

 as Peripatus. 



In remarking upon my statement that, judging from 

 Limulus, the cartilaginous skeleton of the arthropod race, 

 which was dominant when vertebrates first appeared, had 

 arrived both in structure and position exactly at the stage 

 at which the vertebrate cartilaginous skeleton starts, the 

 reviewer states : — " This almost sounds like proving too 

 much, yet it does not account for the vertebrate's dorsal 

 axis." I fail entirely to understand the purport of this 

 remark; there is no cartilaginous dorsal axis in Ammo- 

 ccetes ; he cannot, surely, be thinking of the notochord, 

 which cannot possibly be classed among cartilaginous 

 skeletal tissues. W. H. G.askell. 



" Blowing " Wells. 



In Nature of May 20 Mr. Sydney H. Long describes 

 some " blowing " wells near to Norwich, and intimates 

 that he had not heard of such before. Actually, such 

 wells are not uncommon, and in a recently published 

 memoir of the Geological Survey, on " The Water Supply 

 of Bedfordshire and Northamptonshire from Underground 

 .Sources," some are described (cf. Duston, Long Buckby, 

 Northampton). 



A consideration of the varied phenomena presented by 

 ■" blowing " wells seems to necessitate belief in three 

 possible causes :■ — wind, variations in atmospheric pressure, 

 and fluctuations in water-level. 



Wind can only be effective in very special and obvious 

 circumstances, and so a gusty " blowing " well is a com- 

 paratively rare phenomenon. 



Most water-bearing beds are fed by the slow percolation 

 of water downwards through porous material, and when 

 such a bed is filling up there must of necessity be a dis- 

 placement of air under a pressure greater than the then 

 atmospheric pressure ; indeed, the rate of percolation of 

 water through moderately fine material, such as sand, 

 <leep dow'n in the ground, must be materially retarded by 

 the increasing air pressure. Supposing, however, that a 

 well exists in such a formation, and that the rock is ex- 

 posed, then fluctuations in atmospheric pressure will be 

 mimediately effective in the well, but only after a con- 

 siderable period acting through the water-feeding area, 

 hence every such well will in a sense " blow " when the 

 atmospheric pressure falls. Quite recently I have been 

 interested in a new well being made to the Lower Green- 

 sand ; here, at a depth of about 100 feet, when the baro- 

 meter dropped to 29-3 inches candles were extinguished, 

 and, of course, the men could not work, although at a 

 higher atmospheric pressure no inconvenience was experi- 

 enced. Naturally, at first, the air squeezed out of a deep- 

 seated porous bed is likely to be highly charged with 

 carbonic acid gas, as this was. An old, deep, disused, and 

 covered well a mile or more away from the one just re- 

 ferred to, that had a pipe fixed in the cover, is said by 

 the people living near to give a " trumpeting sound during 

 stormy weather." 



In the case of rocks yielding water abundantly only 

 from fissures, in-draught and out-draught of air from these 

 fissures, in a well, is essentially a question of a falling 

 or rising water-level. When the water-level in such a 

 rock is sinking over a large area, slight though it may 

 he as measured in depth, it draws in, mostly through the 

 fissures, an amount of air equivalent in volume to the water 

 ■being lost by running springs or by pumping elsewhere. .'V 

 rTsing water-level, which mav onlv be obvious in the well 



NO. 2067, VOL. 80] 



long after the rainfall causing it, will, of course, convert 

 such a well into a " blowing " well, with or without a 

 hissing sound depending upon the size of the fissures and 

 the rapidity of rise in the water-level. 



The Drumming Well at Oundle, in Northamptonshire, 

 which was rather noted some 200 years back, no doubt 

 owed its peculiar characteristics to air being forced 

 through a water-lock in the crevices whence the water 

 itself came, with a rising and possibly also a falling 

 water-level. It was sometimes silent for years, and then 

 broke out again, which naturally precludes variations in 

 atmospheric pressure as a cause. 



Northampton. Beeev Thompson. 



Dew-Ponds. 



The article in Nature of April 22 emphasises the fact 

 that the interesting problem of the dew-pond still awaits 

 a definite solution. That these ponds are mostly fed by 

 mist, and not dew, can hardly be doubted by anyone who 

 has visited them at night, situated as they are on the 

 topmost ridges of the Downs. In the driest summer the 

 prevailing south-west wind, as it comes up from the sea, 

 forms on these heights after dark thick clouds of mist, 

 which soak everything that comes in contact with them, 

 and keep green the short grass characteristic of the 

 Downs. 



The source of the water in these ponds, therefore, seems 

 evident, but the mechanism by which the mist is pre- 

 cipitated into the ponds is not so apparent. The question 

 also arises, Why is it essential that the pond be built on 

 the very summit of the ridge of the Downs? Why is 

 it, also, that a few weather-beaten bushes and trees often 

 grow along the ridge of the otherwise bare hills? It 

 appears to me that the only possible explanation is that 

 the particles of mist must bear charges of electricity 

 differing in potential from that of the earth. The charge 

 on the earth would, of course, be most dense at the 

 summits of the hills. Hence the tendency for the mist to 

 deposit on the top of the ridge. 



.About ten years ago I made a rough and somewhat 

 crude experiment to test this theory. The result, which 

 was published in Nature, September 20, iqoo (vol. Ixii., 

 p. 495), was satisfactory so far as it went. Unfortunately, 

 I have never been able to repeat the experiment with 

 better appliances. I feel confident, however, that it is by 

 the investigation of the electrical phenomena of mists that 

 the problem of the dew-pond will be solved. 



Arthur Marshall. 



Naini Tal, India, May 12. 



The Colours of Leaves. 



The notice of Prof. Stahl's book under the heading of 

 " Why Leaves are Green " in Nature of June 3 (p. 393) 

 leads me to direct attention to the effect of protection 

 when applied to our copper beech trees. For the last two 

 years I have, in the spring, partially covered with sacking 

 about half of a small tree (less than 6 feet high), leaving 

 one side open so that there should be some access of light. 

 The aim was to protect a few branches from the effects of 

 frost. This year the cover was put on the part which 

 last year was left uncovered, and about the middle of 

 April, before any leaves had appeared. The cover was 

 removed on May 22 in the presence of several members of 

 the Geologists' Association ; the whole of the sheltered 

 leaves were seen to be quite green, and a remarkable 

 contrast to the others. In two days, however — protection 

 being abandoned — the green leaves commenced to resume 

 their usual spring coloration, and now are, with a few 

 exceptions (as where one leaf may have been shielded by 

 another), of the same tint as the other leaves, and probably 

 no one w-ould suspect they had ever been green. 



The experiment, I suppose, shows the effect of our cold 

 nights in .April and May, which damaged, producing slight 

 chemical change, but did not actuallv kill, the fol'aijp. 

 In a few months' time all the " copper " colour will have 

 disappeared (? been absorbed), and the tree be as green 

 as our common English beech. George Abbott. 



4 Rusthall Park, Tunbridge Wells, June 7. 



