NATURE 



[August 17, 191 1 



should be spelt — which is considered to connect the typical 

 mammoth by means of E. trogontherii with the broad- 

 plated E. meridionalis of the Val d'Arno and Forest-bed. 

 Two molars from Tiraspol are stated to approximate re- 

 spectively to those of E. armeniacus and E. antiquus, but 

 it is scarcely likely that three more or less closely allied 

 forms occur in one deposit. E. trogontlierii is recorded 

 from Nijni-Novgorod, E. meridionalis from Kowialnik, 

 and the typical primigenius from a prehistoric station at 

 Kievo-Kirillovskaia. Finally, a molar from Tiraspol and 

 a second from Bessarabia are respectively compared with 

 those of the Siwalik E. hysudricus and E. planifrons. 



The important part of Madam Pavlow's paper is, how- 

 ever, contained in the discussion as to the mutual relation- 

 ships of the various species and races. After noting the 

 resemblances between hysudricus and meridionalis on 

 one hand and antiquus and namadicus (which some 

 naturalists regard as inseparable) on the other, the author 

 suggests that meridionalis, by an increase in the number 

 and degree of compression of its molar plates, passed by 

 means of wiisti and trogontherii into the mammoth, which 

 died out without descendants. On the other hand, a thin- 

 plated phase of the meridionalis-hysudricus group appears 

 to have given rise to antiquus and namadicus, while the 

 latter in turn produced the modern Indian elephant. The 

 idea that antiquus was the ancestor of the living African 

 elephant is considered improbable. 



The main objection to these views appears to be the 

 phylogenetic separation of the Indian elephant from the 

 mammoth, the two being closely connected by the so- 

 called E. armeniacus, which was probably the animal 

 hunted by Thothmes III. in Mesopotamia. Moreover, the 

 suggestion that E. namadicus ( = antiquus) was the parent 

 of the Indian species is unlikely on account of the peculiar 

 form of the forehead in the extinct species. That the 

 meridionalis-hysudricus line gave origin to the Indian 

 elephant, and that the mammoth branched off from the 

 same stock, perhaps, as Dr. Andrews has suggested, by 

 way of armeniacus, is a far more probable supposition, 

 and one that fits in with all the facts. In regard to the 

 African elephant, there is a general tendency to connect 

 it with antiquus, Dr. Andrews even going so far as to 

 suggest (" Guide to Elephants in Brit. Mus.," p. 42) 

 that the narrow-toothed form of the latter may have been 

 the actual ancestor, or at all events nearly related to the 

 ancestor, of the existing species, although in a previous 

 passage (p. 39) he states that antiquus is unlikely to have 

 given rise to descendants. 



While venturing to dissent in some degree from her 

 theoretical views, I may conclude by expressing apprecia- 

 tion of the value of the work of Madam Pavlow, as it is 

 only by means of such investigations that we can hope 

 to solve the riddle of the elephants. R. L. 



WORK OF THE PHYSIKALISCH-TECHNISCHE 



REICHSANSTALT IN 1910. 

 *T**HE subjoined notes, based upon the annual reporl "I 

 the above institution for last year, indicate a few of 

 the more important researches, &c, undertaken. 



One of the chief researches was the joint work carried 

 out at the Bureau of Standards, Washington, in conjunc- 

 tion with representatives of the English, French, and 

 American standardising laboratories, the most important 

 portion of this work being the determination of the value 

 of the E.M.F. of the Weston normal cell. This was 

 found to be 1-0183 international volts at 20 C. within 

 limits of 1/10,000, agreement 1 >. ing secured in this respect 

 among the countries mentioned. The value given has 

 therefore been accepted in Germanv as from januarv 1 

 last. 



A research on the specific heat of gases al low tempera- 

 tures by the continuous-flow method has been made. In 

 using this method, a measured quantitv of energy C'R is 

 conducted electrically to a gas passing through a tube at 



a constanl r, 1 flow. If the temperature-difference 1" 



flowing and outflowing gas is known when the 



has been attained, as also the quantity of 



Q flowing through the calorimeter in a certain interval 



of time, then > C 2 R is (hfl , , , „ 1( . „ as „_ 



1 Q<# 



so. 2181, vol.. 87 



1 id' i\ no thermal loss takes place, J being the mechanical 

 equivalent of heat. 



In the course of the ordinary conductivity tests on 

 copper carried out during the last few years, it has been 

 found that with great approximation proportionality exists 

 between temperature coefficient and electrical conductivity, 

 i.e. that a very approximate formula was <x 15 .C 15 = o 

 (o ls temperature coefficient, c, s specific resistance in ohms 

 m mm? at 15° C). The mean value for all types of 

 copper tested at the Reichsanstalt since 1905, for the 

 constant, is 6-7 6 -io- s . The same relation seems to hold 

 — of course, with other values for the constants — for 

 aluminium and iron. A similar relation has been found 

 by Dellinger at the American Bureau of Standards. 



The investigation into the variation of wire resistances 

 with atmospheric humidity has been continued, and further 

 experiments made on coils hermetically sealed in accord- 

 ance with the suggestion of the Bureau of Standards. 

 Two coils were filled with petroleum and two with paraffin 

 oil, and sealed up, measurements being made before and 

 after sealing. The coils filled with paraffin oil have shown 

 good constancy, while the petroleum-filled ones have not 

 been so constant. 



A comparison has been carried out between the German 

 standard petroleum testers and four English testers, the 

 result being that the flash-point as given by the English 

 instruments is, on the average, 2° C. lower than with the 

 German instruments, the same oil being used for both. 



Some comparative tests have been made on Seger cones 

 in the electric and the ceramic furnace, the results show- 

 ing that the cones collapse in the ceramic furnace at much 

 lower temperatures than in the electric furnace of the 

 Reichsanstalt. A definite opinion as to the reason for this 

 difference is not pronounced. 



Investigations have been instituted into the change in 

 length of hardened steel. The twenty sets of end rods, of 

 10, 25, 50, and 100 mm. length, forming the basis of the 

 experiments, were again measured in November, 1910. 

 The lengths of the great majority of test-pieces have 

 become constant, four years after manufacture : the 

 changes observed in the remainder are within small limits 

 (fractions of a micron). The results are to be published 

 shortly. 



A series of tests have been made on the energy-loss in 

 dielectrics. An experimental condenser was built up of 

 ten plates of solid insulating material interleaved with 

 copper-foil sheets, the capacity being from 0-004 *° 

 o-oi mfd. A description of the method of testing is given, 

 and the results up to now show that over a range of 

 frequency 9 to 2000 periods the phase-variation in the 

 case of some substances is only to a slight extent dependent 

 on the frequency, while in the case of others thr- varia- 

 tion is considerable. Sometimes it was also noticed that 

 the phase-difference depended on the voltage applied. 



Numerous other researches more or less important in 

 character were undertaken during the year, but space will 

 rot permit of describing them her,'. Those interested will 

 find the reporl of rhe Reichsanstalt published in the Z.it- 

 schrift fur Tnstrumentenkunde for April, May, and June. 



E. S. Hodgson. 



RECENT PUBLICATIONS OF ECONOMIC 

 ENTOMOLOGY. 

 TNSECT pests of trees and crops demand constant atten- 

 •*■ tion on the part of the expert, and a very voluminous 

 literature is growing up round the subject. Few labora- 

 more prolific in published papers than those of 

 Hi. Bureau of Entomology of the United States Depart- 

 ment of Agriculture. Among recent papers, we note one 

 by F. M. Webster on the alfalfa weevil (Phytonomns 

 murinus, Fab. I, a pest introduced from Europe or North 

 Africa some six years ago, and now spreading somewhat 

 widely in Utah, and another paper by the same author 

 on ili-' lesser clover-leaf weevil (P. nigrirostris, Fab.), an 

 insect introduced probably fifty years ago. but not very 

 common even yet ; it suffers from at least two parasites, 

 a small Tachinida and a fungus, Empusa sphaerosperma. 

 The broad-nosed grain weevil (Caulophilus latinasus. Say) 



,1, cribed bj F. H, Chittenden, and also the long-headed 

 flour 1 tli' (Latheticus orytae, Waterh.) ; both are found 



