April 27, 1905] 



NA TURE 



611 



lamp gives 22 candle-power with an energy consump- 

 tion of 1.7 watt per candle-power, or about half that 

 required by the ordinary incandescent lamp. The 

 weight of a single filament is 0022 gram, so that 

 I kilogram of metal would be sufficient for. 45,000 

 such lamps. 



Whether it will be possible to obtain sufficient 

 mineral to produce tantalum on a really large scale 

 remains to be seen, because if it is possible there 

 should be hardly an end to the usefulness of this 

 metal, which possesses the properties ductility and 

 hardness in such an extraordinary degree, leaving 

 entirely out of question its employment in electric 

 lamps. . F. MoLLWo Perkin. 



PRIMITIVE WATER-SUPPLY.' 

 T^HE mighty earthworks that still crown so many 

 ^ of our hills fill the archseologist alike with 

 wonder and despair — wonder that prehistoric man, 

 with the most primitive tools, was equal to the task 

 of raising them, and despair 

 that so little can ever be 

 known about them, despite 

 the most laborious and 

 costly excavation. Plenty of 

 tooks, however, of the kind 

 now under notice would do 

 much to solve the mystery 

 and increase our admiration 

 for Neolithic man, for it is 

 to the period before bronze ' 



was known in Britain that 

 the authors assign the stu- 

 pendous works of Cissbury 

 and Chanctonbury on the 

 South Downs. 



This is an open-air book 

 that gives life to the dry 

 bones of archaeology, and 

 reads like the record of a 

 well-spent holiday. A keen 

 eye for country is one of th.' 

 qualifications possessed by 

 one or both the authors, and 

 «vidence of ramparts long 

 since levelled i« wrung from 

 the very daisias as the\ 

 grow. The construction of 

 dew-ponds by the early in- 

 habitants of Britain has 

 often been glibly asserted, 

 but few, if any, have fur- 

 nished such clear and cir- 

 cumstantial evidence as the 



authors of this short treatise. The water-supply 

 for the occupants of our huge prehistoric " camps " 

 has always been somewhat of a mystery, and it 

 has been suggested that they were only tem- 

 porary refuges, when the country was " up," so 

 that a permanent supply was not regarded as a 

 necessity. But the watering of men and animals on 

 the scale indicated by the areas enclosed would be a 

 formidable task even for a dav, and another explana- 

 tion must be sought. The late General Pitt-Rivers, 

 for example, held that the water-level of the combes 

 was higher then than now, and streams would have 

 been plentiful on the slopes; but, feeling the 

 Inadequacy of this view, he also had recourse to the 

 dew-pond theory. To those familiar with the 

 process, this might seem an obvious expedient, but 

 the interesting account given of the formation of 



such reservoirs might make us chary of crediting 

 prehistoric man with such scientific methods. 



An exposed position innocent of springs was 

 selected, and straw or some other non-conductor of 

 heat spread over the hollowed surface. This was 

 next covered with a thick layer of well puddled clay, 

 which was closely strewn with stones. The pond 

 would gradually fill, and provide a constant supply 

 of pure water, due to condensation during the night 

 of the warm, moist air from the ground on the 

 surface of the cold clay. Evaporation during the day 

 is less rapid than this condensation, and the only 

 danger is that the straw should be sodden by leakage. 

 It is for this reason that springs or drainage from 

 higher ground are avoided, as running water would 

 cut into the clay crust. 



Some ponds of this kind, no doubt of very early and 

 perhaps of Neolithic date, may still be seen in working 

 order : others are of modern construction ; but to and 

 from the ancient dew-ponds (or their sites) can some- 

 times be traced the hillside tracks along which the 



: — Catlle-ways 



' " Neolithic Dew-ponds 

 Hubbard. Pp. x-J-og ; illusi 

 T905 ) Price 3^. 6tf. net. 



nd Cattle-ways." By A. J. Hubbard and G. 

 Lted. (London: Longmans, Green and Co., 



Dew-pond at the North of Cissbury Ring, 

 thic Dew-ponds and Cattle-ways." 



herds were driven, one leading from the camp, or 

 cattle-enclosure hard by, to the watering-place, 

 another leading back, to avoid confusion on the road. 

 These and other details as to guard-houtes and posts 

 of observation are brought to our notice in the 

 description of selected strongholds in Sussex and 

 Dorset; and verification, if, indeed, such is demanded, 

 must be sought on the spot by any who have doubts 

 or rival theories. 



The banks, that enclosed pasture-areas sometimes 

 of vast extent, were no doubt stockaded against man 

 and beast, and may be compared with the base- 

 court defences of the Norman burh ; but the excavator 

 of Wansdyke had an alternative theory that such 

 banks were sometimes erected for driving game. 

 Incidentallv, the authors discountenance the view that 

 the " camps," not to mention the outworks, were 

 ever efficiently manned. Their extent would 

 necessitate for this duty a vast number of fighting- 

 men within call. 



NO. 1852, VOL. 71] 



