424 
iron, zinc and copper mines, on various artesian wells, and on 
the presence of chlorine in certain natural waters. It is recom- 
mended that the State authorities should, conserve all water- 
sheds likely to be drawn upon in future by large towns. 
THOsE who attended the meeting of the Geological Society 
of London on January 8 were impressed with the lucid explana- 
tions of Glacial phenomena in the north of England then given 
by Prof. Percy F. Kendall and Mr. Arthur R. Dwerryhouse. 
Their observations recently published (Quarterly Journal Geol, 
Soc. for August) constitute most valuable contributions to the 
study of the Great Ice Age in this country. Prof. Kendall deals 
with the Cleveland Hills, and points out the evidence there 
existing of the former occurrence of a number of glacier-lakes 
or ‘‘extra-morainic”’ lakes, such as are produced whenever a 
glacier or ice-sheet advances against or across the general slope 
of a country and impounds the natural drainage. Evidence is 
given of such lakes of large and small dimensions in the Vale 
of Pickering, in Glaisdale and Eskdale, in Harwood Dale and 
at Hackness, lakes which must have been formed when the ice 
occupied the Vale of York and extended along the northern and 
eastern borders of the Cleveland area. The evidence is furnished 
by shore scarps, occasional lacustrine deltas with fan-like forms, 
by laminated lacustrine deposits such as the warp clays of the 
vales of Pickering and York, by overflow channels whence the 
impounded water escaped in gorges which trench the main 
watershed or sever spurs independent of the present natural 
drainage, and by crescentic valleys excavated in the face of ‘a 
hill by water flowing round a lobe of ice. The-Glacial deposits 
are, of course, fully considered from sections and from borings, 
some of which were carried out by Prof, Kendall. The assem- 
blages of boulders and rock fragments lead to the conclusion 
that three main ice-movements affected the area—a northern 
from Scotland and Northumberland, a western from Stainmoor 
Pass and the Tees valley, and an eastern from the North Sea 
and Scandinavia. The general sequence which may be inferred 
from a study of the somewhat complicated phenomena is (1) the 
unobstructed passage of the Teesdale glacier to the coast ; (2) 
the arrival of the Scandinavian ice ; and (3) the invasion of the 
Scottish ice. The author finds no signs of the presence of the 
sea in the Cleveland area at any time during the Glacial period. 
Mr. Dwerryhouse describes the glaciation of Teesdale, Wear- 
dale and the Tyne valley, a region in which, like that of 
Cleveland, the higher tracts stood out as ‘ nunatakkr,” while 
the grounds beneath were buried by ice. He also points out 
that at the period of maximum glaciation a number of lakes 
were formed, owing to the obstruction of the drainage of lateral 
tributary-valleys by the ice of the main glaciers. 
Drs. H. M. HILLeR and W. H. Furness have privately 
issued bound copies of the ‘* Notes of a Trip to the Veddahs of 
Ceylon,” which were published in vol. iii. of the Bud/etin of the 
Free Museum of Science and Art, Philadelphia (April, 1901). 
The ‘* Notes” do not contain much that is new, but they are 
illustrated with several excellent photographs, the most inter- 
esting of which are those illustrating a Rock Veddah shooting 
(standing) with a bow and arrow, and one making fire with a 
“* fire-drill.” 
THERE is in Zhe Reliqguary and Illustrated Archaeologist 
(vol. viii. No. 3) a well-illustrated article, by Mr. R. Quick, on 
the Carib stone implements in the Horniman Museum, These 
implements show the technical skill of the aborigines of the 
West Indies in working hard rock. Some of the implements 
are really remarkable examples of stone-work. One example 
which is figured has a most irregular contour ; from its high 
finish it was. evidently greatly prized, and was probably a 
symbolic religious object, of which the significance is at present 
unknown. 
NO. 1713, VOL. 66] 
NATURE 
[Aucust 28, 1902 
The early Christian monuments of the Isle of Man are 
becoming well known to students through the enthusiastic 
labours of Mr. P. M. C. Kermode. In the July number of 
The Reliquary and Illustrated Archaeologist he gives numerous 
illustrations of recently discovered crosses and runic and ogam 
inscriptions. One of the most interesting discoveries is a carved 
stone with very characteristic Scandinavian interlacing ; on one 
side is seen the figure of Loki in the act of heaving stones at 
the otter which is eating the salmon it has just caught in the 
foss ; above this is the steed Grani with the chest containing the 
hoard won by Sigurd upon his slaying the dragon Fafni, 
MEssrs. PERKEN, SON and Co. have issued a new revised 
and enlarged edition (the eighth) of their ‘‘ Beginner’s Guide 
to Photography.” The book is now in its seventieth thousand. 
IN the article on ‘‘ The Older Civilisation of Greece,” which 
appeared in NaTureE of August 21, the following corrections 
should be made :—P. 391, co/. 1, 7. 9 from bottom: for ‘*the 
un-Aryan ‘ Pelasgian’” read ‘‘the probably un-Aryan 
‘Minoan’”; p. 393, cod. 2, 2 12 from top: for ‘‘ ev wux@, 
“Apyeos” vead ‘‘ ev wux@ ~Apyeos” ; zbid., 2. 14 from top: for 
‘*casements” read ‘‘casemates”; p. 394, col. 1, Fig. 2: for 
*€Clay Seal. Impression” vead ‘‘ Clay Seal-impression.”’ 
A MINUTE investigation of the composition of Pennsylvania 
petroleum has recently been carried out by Mr. C. F. Mabery, 
and an account of the hydrocarbons with boiling points above 
216° C, is published in the Proceedings of the American Academy 
of Arts and Sciences (vol. xxxvii. p. 565). Hydrocarbons of the 
methane series from tridecane Cj,H,, to octocosane CogH;, have 
been isolated, and according to the molecular-weight determina- 
tions carried out by the freezing-point method the products 
obtained by the author as the result of repeated fractional dis- 
tillation under low pressure are almost pure. No account 
appears to have been taken, however, of the possibility of the 
occurrence of different isomeric forms. In addition to these 
saturated compounds the hydrocarbons Cy,.H 44, Cogi Tyg, CoyHys, 
CH; and C,-H;, belonging to the ethylene series and C..H,;4 
a homologue of acetylene, have been obtained. 
MUCH remains to be learnt about the numerical relationships 
of the atomic weights of the elements. It has long been 
realised that when referred to the standard O = 16 many of the 
atomic weights approach whole numbers to an extent out of all 
proportion to the probabilities of the case. In the Chemzker- 
Zeitung for July 19, Mr.- Arthur Marshall, as well as directing 
attention to this fact, shows that very remarkable relationships 
exist in many cases between the atomic weights of allied elements. 
Taking from the tables of the German Chemical Society the 
eighteen values given to two places of decimals, the theory of 
- probabilities shows that the chances against their approaching as 
close as they do to whole numbers are as high as 4120:1. If, 
on the other hand, the atomic weights are referred to H = 1, 
there appears to be little or no tendency to become whole 
numbers. It is, however, only when certain of the atomic 
weights are referred to entirely different standards that the most 
striking relationships appear. The weights of the atoms of the 
halogen elements and silver, for instance, are exactly in the ratio 
Cl: Br: Ag: I = 90: 203: 274: 322. In the case of the alkali 
metals the proportions are even simpler, Li: NH,: Na: K: Rb= 
7:18:23:39:85. Again, the horizontal series, V:Cr: Mn: 
Fe: Ni:Cu:Zn = 54:55:58:59:62:67:69. It is yet pre- 
mature to work out relationships for all the elements, for there 
is still great uncertainty about most of the atomic weights, but 
the values for most of the above substances are thoroughly well 
established. 
MEssrs. HABER AND GEIPERT have been investigating the 
conditions under which aluminium is obtained by the electrolytic 
method, and have published their results in recent issues of the 
