F. J. Bennett — Machine-made Implements. 69 



Fig. 7. AT^ertwxe oi SemimuUelea acupimctata i^OYak) . After Novak. 



8. Ditto. 



9. Aperture of RcptomuUelea sarissata, Gregory. From the type-specimen 

 (B.M. D7106). 



10. Aperture of Rej^tomultelealteussi (Pergens). After Pergens. Other zoascia 

 in Pergens' figure have a much more rounded distal angle. 



11. Apertm-e of Reptomultelea tuber osa, d'Orbigny. From the specimen 

 B.M. 36,746. 



12. Part of the zoarium of the type-specimen of Semimultelea Dixoni, n.sp. 

 (B.M. i) 7845), showing normal zooecia {n.z.) ; closed zofficia (e.z.) ; 

 a topomorph with ridges between the zooecia {t.r.) ; and a topomoqih 

 composed of areas marked out on the zoarium by raised ridges, but with 

 no closed or open apertm-es (<;.). x about 24. Somewhat diagrammatic. 



V. — Machine-made Implements. 

 By F. J. Bennett, F.G.S. 



ATTENTION has lately been called by M. Marcellin Boule to 

 the production in cement-mills in the Commune of Guerville, 

 near Mantes, of air the more characteristic forms of Eoliths, and 

 of these he has given photographic reproductions. The evidence 

 for the necessarily artificial shaping of Eoliths had for many years 

 been questioned by him, because he had found chipped flints of this 

 character in the midst of Oligocene or Miocene beds in Auvergne 

 and in the Velay ; and it seemed imprudent to infer the existence 

 of man in those early stages of the Tertiary period in the absence of 

 osteological evidence. 



In speaking here of machine-made implements I do so advisedly, 

 because all stone implements were once referred to natural or 

 supernatural causes ; the obvious arrovf-head, for instance, being 

 termed an ' elf-bolt.' 



M. Boule, however, seeks only to show that stones shaped like 

 Eoliths may be produced by Nature, because he finds that they are 

 produced by certain pseudo-natural, machine-made torrents, and so 

 considers that Eoliths are due to such torrential action. Yet the 

 Eolithic deposits known to the writer do not seem to indicate 

 torrential action. 



The first objection is that M. Boule compares known and 

 unnatural agencies with natural ones, and the analogue of his 

 machine-made torrent would be hard to find in Nature and would be 

 most exceptional there, and j'et to this he v/ould refer all Eoliths.^ 

 The Mantes wash-mill apparently deals only with flints fresh from 

 the chalk, while the flints from which the Kent plateau Eoliths 

 were made were mostly tough and much weathered, and not as a rule 

 such flints from which good chipping can be obtained ; and that 

 may account a good deal for their rough execution, and for their 

 non-acceptance by some observers. 



Anyone who is a flint-knapper knows that the results obtained 

 from the one kind of flint are very different from those obtained 

 from the other, that the fracture varies with the flint, and that in 



1 Nature, Aug. 31, 1905, p. 438 ; Sept. 28, p. 538 ; and Oct. 26, p. 635 ; see 

 also L'AnthropGloffie, vol. svi, p. 257, in Avhich the detailed observations of 

 M. Boule are published. 



