512 



NATURE 



[December 15, 192 1 



The Mound-builders of Dunstable. 



AT a meeting of the Roval Anthropological Insti- 

 tute, held on November 8, Dr. W. H. R. 

 Rivers, president, in the chair, Prof. G. Elliot Smith 

 and Capt. Guy Crowden read a paper on "The Mound 

 Builders of Dunstable." After describing tW results 

 of excavations on one of the Five Knolls on Dun- 

 stable Downs, in which the remains of three cremated 

 bodies were interred, probably in the Bronze age, the 

 authors directed attention to the association of the 

 tumuli with cultivation terraces, huts, and ancient 

 roads, and suggested that the presence of flint suitable 

 for implement making was the determining cause of 

 the settlement of the people who built the huts and 

 made the cultivation terraces on the Dunstable 

 Downs. The convergence of the main roadways at 

 this spot is also to be explained by the transport of 

 the most valuable economic product of the Neolithic — 

 and even also of the Bronze — age to places where 

 such material could not be obtained locally. Atten- 

 tion was directed to the geographical distribution of 

 cultivation terraces in Britain, and their remarkable 

 association in so many places with the edge of the 

 chalk ; and the attempt was made to correlate these 

 facts with the observations of Mr. W. J. Perry as to 

 the causal relationship between the distribution of the 

 megalithic monuments of Wiltshire, etc., and the flint- 

 bearing edge of the same chalk zone further south. 

 The plea was made for the fuller investigation of the 

 relationship existing between ancient monuments and 

 geological formations that produced substances valued 

 by man in ancient times, and also for the investiga- 

 tion of the effects of the admixture of cultures re- 

 vealed in the round barrows in different parts of the 

 country. 



The discussion which followed the reading of the 

 paper dealt mainly with the question of terrace culti- 

 vation. The authors in the course of their paper had 

 suggested that the employment of terrace cultivation 

 on this site was due to a conservative instinct which 

 continued to employ a traditional system of cultivation, 

 originating elsewhere, and not necessitated by the 

 conditions of the present site. Mr. Peake, however, 

 suggested that terrace cultivation was a natural con- 

 sequence of ploughing or hoeing on the side of a hill : 

 these operations, in course of years, would be bound to 

 produce terraces such as those known on the Downs 

 as "shepherds' steps." Col. Hodson, on the other 

 hand, pointed out that in Assam, also an area^ in 

 which megalithic monuments occurred, terrace cultiva- 

 tion was practised by the Nagas, and the terraces, so 

 far from being the result of the method of cultivation, 



were built up of set purpose when land was brought 

 under cultivation. Mr. Mills also stated that thf 

 Sema Nagas, when urged by the administration to 

 follow the terrace system, had staled that they were 

 unable to do so, as they did not know the sacrifices 

 for terrace cultivation. Mr. Strong said that in China 

 terrace cultivation had been brought about by climatir 

 changes. Owing to deforestation the climate had 

 changed, and it had been necessary to introduce ter- 

 races with shale retaining walls. 



Mr. Peake, in the course of his remarks, also dealt 

 with the question of the roads which meet at or near 

 Dunstable. He pointed out that the Icknield Way fol- 

 lowed the junction of the chalk and the greensand, 

 and suggested that while its course was determined 

 by the position of the springs which were found at 

 that junction, the course of Watling Street was deter- 

 mined by purely geographical conditions, and depended 

 upon the position of the Dunstable gap. The site 

 of the Dunstable settlement had been determined by 

 the roads rather than vice versa. As regards the 

 origin of megalithic buildings, he had begun to think 

 we must look further east than Egypt, possibly in the 

 Persian Gulf. A note recently published by Prof. 

 Sa^xe in Man showed that the people of Akkad were 

 interested in a tin-land at a very early date. It was 

 possible, therefore, that the megalithic people were 

 not Armenoids. The second stream of broad-headed 

 invaders of this country were the "Beaker-folk," the 

 centre of distribution of whose culture appeared to be 

 Bohemia, or possibly Southern Russia, but it showed 

 no trace of ^gean influence. He did not regard the 

 authors' correlation of chalk and megaliths as con- 

 vincing if the distribution in this country were taken 

 as a whole. Mr. Crawford had suggested a more 

 reasonable explanation of early settlement in pointing 

 out that it depended upon the distribution of grassland 

 and forest area. Chalk and limestone gave grassland 

 areas, and were, therefore, the earliest to be peopled. 

 Mr. Garfitt pointed out that the stone circles of Derby- 

 shire did not comply with conditions suggested by the 

 authors. 



Prof. Elliot Smith replied briefly to his critics. He 

 maintained that the position of the roads was deter- 

 mined bv the occurrence of siiiiahle flint, which was 

 not found at anv and every point ; he failed to under- 

 stand how a people such as the Elamites, who used 

 brick and built no megalithic monuments, could have 

 been responsible for the diffusion of the megalith ; and 

 pointed out that the Derbyshire stone circles were 

 associated with copper. 



Norwegian Meteorology. 



(i) A PUBLICATION entitled " Nedboriakttagel- 

 •'*- ser i Norge, " recently received from the 

 Norwegian Meteorological Institute, contains informa- 

 tion relative to precipitation at about five hundred Nor- 

 wegian stations over periods of from lo to 40 years 

 ending 1915. The tables give mean and extreme 

 monthly and annual values, as well as frequencies 

 of occurrence of the various types of precipita- 

 tion, while the charts shew the geographical 

 distribution of some of the tabulated elements. 

 It wi'l be seen that the days on which pre- 

 cipitation of i/io mm. or more is measured have a 

 frequency of from 50 to 60 per cent, (per annum) in 

 the western coastal regions, falling to 30-40 per 

 cent, in the more easterlv districts cf the south and 

 inland districts of the north. Great variability occurs, 



NO. 2720, VOL. 108] 



however, over comparativelv small areas. Whereas 

 Osland (Bergen) reports i mm. or more on an aver- 

 age of 197 davs in the year, Ulstad, which is also in 

 South Norway, but further east, records this amount 

 on onlv 54 davs. ^ ^^ 



The introduction to " Nedberiakttagelser i Norge 

 is devoted mainly to a discussion of wind screens for 

 rain gauges. The results of experiments show a general 

 increase in the amount of precipitation measured when 

 screens are in use, especially in winter. 



(2) " Om veir og vind i Trondhjem." In this paper, 

 which runs to about seventy pages, M. K. Hakonson 

 Hansen summarises and discusses meteorological ob- 

 servations at Trondhjem during -^o^ years {1885- 

 1915), and presents numerous tables of mean and 

 extreme values, including, among other things, an 



