August 6, 1908] 



NA TURE 



329 



Sussex. The name usually given to these remarkable 

 structures probably connects them in popular tradition with 

 the Danes. The writer reviews the many theories which 

 have been advanced to account for their origin — that they 

 were flint-workings; places of secret worship and reposi- 

 tories for the remains of the dead ; underground dwellings 

 or hiding-places; excavations for the extraction of chalk. 

 All these are found to be inadequate. The amount of 

 flint which could have been obtained in such places is in- 

 significant, and no attempt seems to have been made to 

 work the small e.xisting supplies ; there is no evidence that 

 they served any purpose in connection with a religious 

 cultus ; the existence of many excavations within a limited 

 area renders it improbable that they were dwelling-places 

 or refuges, their contiguity exposing them to simultaneous 

 hostile attack, while their construction offers no facilities 

 for escape ; their character shows that the removal of chalk 

 was not the primary intention. Mr. Philip favours the sup- 

 position that they were intended to be used as silos, or 

 imderground granaries, and most of the arguments adduced 

 in opposition to the other theories contribute to establish 

 this view. Thus they were naturally close together, because 

 the tribe clustered within a narrow area. The principal 

 objection is the risk of damp ; but, as a matter of fact, 

 the caves seldom show signs of being affected by under- 

 ground moisture, and perishable grain was probably pro- 

 tected by thick surrounding layers of straw. That they 

 were not the work of a single age is shown by the fact 

 that the character of the pick-marks varies ; in the older 

 specimens we find marks of horn, bone, and flint tools ; 

 in those of a later date metal implements were used. The 

 shaft is invariably some 3 feet in diameter, and seldom 

 less than 60 feet in depth, ingress and egress being pro- 

 vided by means of a rude ladder or by ropes of hide. 



The Deutsche Seewarte has recently published a double 

 number (Heft 15 and 16) of its collection of over-sea 

 meteorological observations, referring chiefly to the years 

 1905-6. The work includes (i) monthly and yearly means 

 of eye observations and automatic records made under the 

 supervision of the Seewarte at twenty-five stations ; (2) in- 

 dividual observations and hourly means at a number of 

 selected stations in German East Africa. The whole of 

 the tables give evidence of great care in their preparation 

 for publication, to the cost of which the Imperial Colonial 

 Office contributes ; they furnish a very valuable contribu- 

 tion to the meteorology of various remote parts of the 

 globe. We are glad to note that, in cases where only 

 summaries are printed, the original documents are avail- 

 able on loan for the purpose of scientific investigations 

 under reasonable restrictions as to their safe return. 



Five years ago M. Ivar Frcdholm published in the Ada 

 Mathematica a remarkable memoir which seems likely to 

 prove the starting point of a host of important results, both 

 in function-theory and in mathematical physics. One of 

 the latest works to which it has given rise is the thesis 

 of Dr. H. Bryon Heywood, " Sur I'Equation fonctionelle 

 de Fredholm et quelques-unes de ses Applications " 

 (Gauthier-Villars, iqo8J. In this the author gives a 

 summary of the results of Fredholm, Picard, Poincar^, and 

 others, and gives applications to heat-conduction, tides, 

 potential, &c. The first chapter contains an extension of 

 some propositions of Hilbert's ; this appears to be the most 

 original part of the thesis, but the whole is worth reading, 

 because it brings a variety of problems under the same 

 treatment. 



The Physical Kcvicw for June contains a communication 

 from .Messrs. A. H. Taylor and E. H. Williams, of the 



NO. 2023, VOL. 78] 



University of Wisconsin, on the distributed capacity of 

 resistance boxes. They find by the modified bridge method 

 used by Messrs. Rosa and Grover for comparing induct- 

 ances, that the capacity of the ordinary Wolff looo-ohm coil 

 is O-00O5-O-OO06 microfarad, while that of a looo-ohm 

 coil wound on a wood core is only 000007 rnicrofarad. 

 Wolff 500-ohm coils give o-ooo 16-0-0002 2 microfarad, and 

 2000-ohm coils 000041-000063 microfarad. The authors 

 attribute the high capacities of Wolff coils mainly to the 

 comparatively large wire used in constructing them, but 

 also to the shellac between the wire and the brass tube 

 on which it is wound not being thoroughly dry. It seems 

 evident from these results that resistance coils must be 

 used with great discrimination in accurate alternating 

 current measurements. 



In May last M. C. Firy gave before the Societe fran^aise 

 de Physique an account of the methods he had in succession . 

 adopted, in his endeavour to produce a pendulum electrically 

 driven, which should be accurate to within a second a 

 day. His communication is printed in full in the July 

 number of the journal de Physique, and from it we gather 

 that the arrangement he finds most satisfactory is one in 

 which the pendulum carries beneath the bob a horse-shoe 

 magnet, the plane of which coincides with the plane of 

 swing of the pendulum. The lower of the limbs of the 

 magnet enters during its swing a fixed coil, and the upper 

 a small copper ring which forms the bob of a second 

 pendulum having the same time of swing as the first. 

 This ring is carried along with the moving magnet owing 

 to the currents induced in it, and makes or breaks the 

 circuit of the fixed coil, through which a current from a 

 cell in series with it flows in consequence in such a direc- 

 tion as to supply the small impulse necessary to com- 

 pensate for the decay of swing of the principal pendulum 

 owing to friction. The same arrangement is adopted by 

 M. Fery in controlling a number of distant pendulums 

 by means of a central standard. 



Whes sufliciently fine solid particles are suspended in 

 a liquid they exhibit under a high-power microscope rapid 

 motions which are known, from their discoverer, as 

 " Brownian." These movements form the subject of an 

 interesting study by Dr. M. Seddig in the Physikalische 

 Zeitschrift for July 15. The particles were of cinnabar, 

 a quantity of which was stirred up with the liquid and 

 allowed to stand in a long tube for a week. The top 

 layer of liquid was then decanted, and a drop of it 

 arranged as a slide in a microscope fitted with a camera. 

 .'\n arc light provided the strong illumination necessary. 

 By means of a falling shutter with two holes in it, two 

 exposures, each of 1/40 second and i/io second apart, 

 were made, and the motions of a number of the particles 

 between the two exposures measured on enlargements of 

 the original plates. As the result of observations made on 

 the same liquid at different temperatures, Dr. Seddig comes 

 to the conclusion that the theory of these movements given 

 by Einstein, according to which they should vary in 

 intensity as the square root of the absolute temperature, 

 is in keeping with the facts. 



Dr. F. a. Bather proposes to publish a double index 

 to the generic and specific names in E. Desor's " Synopsis 

 des Echinides Fossiles," a work still in constant use by 

 every worker on the Echinoidea. This index will be pre- 

 ceded by a " Note sur les Dates de Publication," drawn 

 up by Mr. Jules Lambert. Intending subscribers to the 

 index should communicate with Dr. Bather at the Natural 

 History Museum, South Kensington, S.W- 



