Jan. 30, 1879] 



NATURE 



303 



upwards of half a mile, but the actual length of wire which was 

 submei^ed, and through which the current passed, was upwards 

 of a mile. The spark was about \ inch in length, bluish white 

 in colour, and very striking and interesting in effect. It was 

 placed in the focus of a holophotalised parabolic reflector. 



At the same time in order to ascertain if by means of a single 

 battery under the charge of one keeper a succession of flashes 

 could be produced and a string of isolated dangers illuaiinated, 

 the light was sub-divided first into two separate flashes and 

 afterwards into six different sparks. The separate lights were 

 quite satisfactory, though they were not as might have been ex- 

 pected of the same power as the original single one. But as the 

 separate sparks were very close to each other this cannot, I pre- 

 sume, be held to be a proper sub-division of the light. 



In 1867, at the British Association at Dundee, I suggested 

 that the " effect of the light might be also increased without using 

 additional cells if the same current could be again utilised so as 

 to generate a second spark in the same focu^. This was pro- 

 posed to be done by ' using additional coils ' for the same focus 

 or separate sparks in the foci of separate reflectors.^ I also 

 added that the "time is perhaps not far distant when the 

 beacons and buoys in such a navigation as the entrance to Liver- 

 pool may be lit up by submarine conduction from a central 

 station on either shore, while the whole management may be 

 trusted to the charge of one or two light-keepers." ^ 



I may add that similar trials were made with Wilde's electro- 

 magnetic machine, which gave a light of much greater volume 

 and power. The electrodes employed in all the;e experiments 

 were made of platinum, but several other metals were experi- 

 mented with, and of all that were tried bisnuth was found to 

 give the brightest light.^ 



A committee of the Scottish Society of Arts, consisting of Dr. 

 Ferguson, convener, Dr. Lees, and the late Dr. Strethili Wright 

 reported in the following lerms : — "The peculiar character of 

 the light, which is flickering, though continuous, is well marked 

 and would be easily understood. So far as Mr. Stevenson's 

 experiments go, they seem to prove the practicability of his pro- 

 posal, and your committee do not anticipate any serious obstacle 

 to its realisation." Thomas Stevenson 



Edinburgh 



UNDERGRO UND TEMPERA TURE * 



T~~iR. STAPFF has continued his observations of the tempera- 

 ^-^ ture in the St. Gothard Tunnel, and has contributed to 

 the Swiss Natural History Society a paper ^ of fifty-six quarto 

 pages, embodying the results. 



The following is his description (pp. 26, 27), of the mode ot 

 observing the temperature of the rocks in the tunnel : — 



"The exact determination of the temperature of the rocks in 

 the tunnel formerly occasioned a notable expenditure of time and 

 money. At first thermometers about a metre long (made by J. 

 Goldschmid, of Zurich) were employed for this purpose ; their 

 tubes being cemented into a wooden cylinder, so that only the 

 bulb (surrounded by a perforated steel cap) projected below, and 

 the scale (extending from 15" to 30° C.) above. Tallow was 

 poured round the wooden cylinder, and the whole thermometer 

 was then thrust into a bore-hole a metre deep, so that only the 

 scale projected, from which readings were taken from time to 

 time until the temperature became constant. The final reading 

 had to be corrected not only for rise of zero but also for the 

 temperature of the quicksilver in the thermometer tube which 

 extends from the opening to the bottom of the bore-hole. 

 Another very notable correciion was required for the more or less 

 oblique position of the thermometer ; for -the hydrostatic pres- 

 sure of the quicksilver presses out the glass bulb so far that with- 

 out change of temperature the long thermometer reads from cP'^ 

 to i°"o less in the vertical than in the horizontal position. 



' ' After about from three to ten days, the reading of a ther- 

 mometer luted into a bore-hole ceased to alter. 



"Separate trials with thermometers of similar construction, 

 but different length, showed, moreover, that, after months, the 

 temperature of the rock at about a metre deep was still un- 



» British Association Reports, 1S67. 2 /HJ, 



3 " Proposals f..r the Illumination of Beacons and Buo\-s." dd ii-i« 

 (Edinburgh: A. and C. Black.) ' ^^' * '' 



* Eleventh Report of the Bntish Association Underground Temperature 

 Committee, by Prof. Everett. 



5 " Studien fiber die Warmevertheilung ira Gotthard," i Theil. " Der 

 5chweiierischen naturforschenden Geselkchaft lu ihrer sechzigsten Tahres- 

 versammlung in Bex gewldmet," von F. M. Stapff. Bern, 1877 



changed. This is obviously owing to the small difference of 

 temperature between the rock and the surrounding air. 



" From the observations at No. 8 and No. 15, in Table III., 

 it is seen that the temperature at the bottom of the bore-hole 

 was sometimes a little lower and sometimes a little higher than 

 nearer its mouth. 



" This mode of observing gave correct results, but was labo- 

 rious and costly, not only on account of the necessity of making 

 special bore-holes for the purpose, but because almost every 

 experiment cost a thermometer. The projecting end was often 

 maliciously broken off, and on account of the swelling of the 

 wooden case it almost never happened that at the end of an 

 experiment a thermometer was drawn out again uninjured. 



" Hermann and Pfister remedied this latter evil by surrounding 

 the thermometer- tube from the bulb to the scale, with a glass 

 case, and this with a steel jacket. This arrangement, however, 

 involves not only conduction through the steel, but also continual 

 interchange of heat by currents of air in the glass case, from the 

 mouth to the bottom of the hole. For these reasons the obser- 

 vations made with these thermometers could not be employed 

 without intricate corrections. 



"Later I tried a Thomson's maximum thermometer,^ kindly 

 placed at my disposal by Prof. Everett, which (after previous 

 strong cooling) was left for several days at the bottom of the 

 bore-hole, closed air-tight. The results agreed with those 

 obtained by other methods ; but who can guarantee that the 

 higher temperature prevailing in a newly-bored hole is always 

 just so much depressed by the cold mass of the thermometer and 

 its copper case, that the rock-temperature alone determines the 

 final indication of the maximum Uiermometer ? 



"This consideration induced me to employ for rock-tempera- 

 ture observations (and they also serve for air and water observa- 

 tions), the above-mentioned short thermometers with instdated 

 bulbs, the first of which Prof. Everett caused to be made by 

 Negretti and Zambra for this express purpose. These thermo- 

 meters, inclosed in a metal box provided with a handle, are 

 thrust to the bottom of the bore-hole, which is at least a metre 

 deep. To the handle is fastened a strong cord reaching to the 

 mouth of the hole, by which it can be drawn out again at the 

 end of the trial. The borehole, from the thermometer to the 

 mouth, is stopped with greased rag or other similar material, as 

 air-tight as possible. After two or three days, the thermometers 

 have usually assumed the temperature of the surrounding rock, 

 that is to say, their reading has ceased to alter. The insulation 

 of the quicksilver prevents alterations during the drawing-out 

 and reading of the thermometer. The correctness of the result 

 is in no way prejudiced by sediment from the boring which may 

 yet remain in the hole. The pouring in of some water may even 

 be useful in accelerating the experiment. Wet bore-holes with 

 standing-water are, however, to be avoided, because rock- 

 temperature and water-temperature are not identical, 



" In the manner last described, at every available opportunity, 

 that is to say, when the work of the tunnel is from any cause 

 compelled to cease for a few days, rock-temperature observa- 

 tions are now instituted in bore-holes ready to our hand. The 

 observations are simple, give exact results if taken with proper 

 precaution and sufficient duration of the experiment, and cause 

 no further expense, since the thermometers, being sunk in the 

 rock, are secured against wanton inji'ry, and there are always 

 bore-holes available." 



Dr. Stapff further states by letter that the two original ther- 

 mometers supplied by Negretti and Zambra having been broken, 

 he has had others made, in which he has intrc^uced the im- 

 provement of hermetically sealing the outer glass case, instead 

 of closing it with a waxed cork, which gradually admitted 

 moisture. 



In the Report for 1876 an account was given of the observa- 

 tions of Herr Dunker in a bore about 4,000 feet deep at Speren- 

 berg, and allusion was made to the undue weight which had 

 been attached by some m riters to the empirical formula in which 

 Herr Dunker sums up his observations ; a formula which indi- 

 cates a retarded rate of increase, and, if extended to greater 

 depths, leads to the conclusion that the temperature reaches its 

 maximum at the depth of about a mile. 



A discussion has been carried on in Germany on this subject,* 



' It was one of the' protected N'egretti maximum thennometers constructed 

 for the Committee. 



^ See papers by Mohr, Heinrich (two papers), Dunker, and Hottetu-oth, 

 in the " Neues Jahrbuch " for 1875, 1876, and 1877: by Braans, in the 

 " Zeltschrift fur die ge^ammten Naturwissenschaf ten, " 1874, p. 483; and by 

 Hann, in the "Zeitschrift der osterreichischenj Gesellschaft fflr Meteoro- 

 logie," 1878, p. 17. 



