240 



SCIENCE. 



[N. S. Vol. II. No. 34 



greater absorption when its fibres were par- 

 allel than when they were perpendicular to 

 the plane of the vibrations. That is, the 

 wood is double refracting and transmits 

 best when the plane of vibration is perpen- 

 dicular to the direction of the fibres and 

 absorbs most when they are parallel. In 

 fact, to this absorption the author ascribes 

 the reappearance of the sparks with the 

 crossed sender and receiver, since the plates 

 of wood were not thick enough to give a 

 retardation of one ray behind the other 

 equivalent to a half wave-length; in fact, 

 the plates were only from {^ to |^ a wave- 

 length thick. The phenomenon was ob- 

 served with plates up to 35 cm. thick and of 

 fir, oak and beech. Plates cut perpendic- 

 ular to the fibre showed no double refrac- 

 tion as was anticipated. The author refers 

 to the experiments of Starkl showing that 

 the thermal conductivity of wood is different, 

 parallel with and perpendicular to the fibre. 



W. von Bezold calls attention to the ex- 

 periments on thermal conductivity by 

 Tyndall, which led him in 1871 to experi- 

 ment upon the Lichtenberg figures formed 

 upon wood. At that time von Bezold found 

 that the Lichtenberg figures on plates of 

 wood cut with the gram fibre were elliptical 

 with the major axis at right angles to the 

 fibre, while the ellipses obtained upon the 

 same wood by Senarmont's methed of testing 

 thermal conductivity were much more elon- 

 gated and in the direction of the fibre. The 

 author was able to give to a hard rubber 

 plate an anisotropic character by pasting 

 parallel strips of tinfoil upon the back, 

 whereupon it gave Lichtenberg figures simi- 

 lar to those on a piece of wood cut with the 

 grain. He suggests experiments with a di- 

 electric in which are imbedded rods of a 

 conductor or a dielectric of difierent in- 

 ductive capacity. 



A. Eighi points out that he had presented 

 a paper before the Academy of Sciences at 

 Bologna, which Mr. Mack evidently had not 



seen, although it was read nearly six months 

 before his article appeared. On that occa- 

 sion Mr. Eighi described experiments simi- 

 lar to those of Mack, wherein he obtained 

 identical results, even identifying elliptical 

 and circular polarization in wood. 



In conclusion, I may state that I have re- 

 cently examined thin films of wood between 

 crossed Nicols, using sunlight, and found in 

 all cases the behavior toward light the same 

 as that described for electric waves, albeit, 

 one sees that the double refraction is not 

 shared equally by all the components of the 

 wood. William Hallock. 



Distribution of the Magnetic Declination in 

 Alaska and Adjacent Waters for the Year 

 1895. Bull. No. 34, U. S. Coast and 

 Geodetic Survey. 8°, 6 pp. and 1 chart. 

 The above is a brief abstract of a fuller 

 report to be published later by C. A. 

 Schott, Assistant. As the title indicates, 

 the paper attempts to give the latest repre- 

 sentation of the distribution of the mag- 

 netic declination for the region indicated. 

 It hence replaces the earlier (1890) at- 

 tempt and gives evidence of a decided 

 improvement. On account of the wide 

 extent of the region involved and the 

 paucity and irregular distribution of the ob- 

 servations; the analytical method of repre- 

 senting the available declinations was 

 adopted. The formula established gives a 

 very satisfactory accord with the observa- 

 tions covering a territory from 47°. 4 to 

 71°.3 N and from 122°.4 to 156°. 5 W of 

 Gr. The largest discrepancy between the 

 observed and computed declination is but 

 0°.9; the probable error of a single repre- 

 sentation, ±19'. 



On the chart the isogonics or lines of 

 equal magnetic declination, as obtained with 

 the aid of the formula, are given at intervals 

 of one degree for the region covered by the 



above title. 



L. A. Bauee. 



