Oct. 3, 1889] 



NATURE 



559 



uncertninties of these measures and those resulting from work 

 with modern instrument?, and following more recent methods. 

 It was shown that the probable err.)rs were much lar^^er than 

 would be indicated by the agreement of the published results, 

 and that therefore the concordance of this arc with those in 

 other parts of the world is no proof of its accuracy. By far the 

 most significant errors came from the astronomical observations, 

 and the unavoidable uncertainties here, either from the imper- 

 fection of the instruments, or the attraction of the high moun- 

 tains, were shown to affect perceptibly the present value of the 

 earth's axis. 



The resolution touching this matter, which was unanimously 

 adopted at a general session of the Association, reads as 

 follows : — 



" Whereas the history of geodesy includes no more important 

 page than that relating to the measurement in 1739 of the so- 

 called Peruvian arc, which work was conducted by the French 

 Government wiih the co-operation of Spanish officers, and in 

 magnitude of plan and difficulty of a most serious character it 

 was in its time unexcelled ; whereas recent improvements in 

 all the processes incidental to such an undertaking have been so 

 vet}' great, rtndering possible a vastly more accurate execution 

 of the work ; wherea-; it is and has been for several years a 

 matter of deep 1 egret that the one great contribution which the 

 American coniinent has made to the solution of the problem of 

 the figure of the earth should fall so short of what it should and 

 might : be it therefore resolved by the Ameiif-an Association for 

 the Advancement of Science, that the Congress of the Three 

 Americas about to assemble in Washington is earnestly requested 

 to consider the desirability of undertaking ihe measurement of 

 this Peruvian arc, to be accomplished by a union of the Republics 

 represented. This result is not likely to be reached except 

 through international effort, and this recommendation by the 

 Congress would be a fitting and proper act of this first confer- 

 ence of representatives of the New World." 



Section rf Physics. 



The President of the Association, Mr. T. C. Mendenhall, 

 read before the members of this Section a short paper on globular 

 lightning. He a Iduced no new theory as to its source or charac- 

 ter, but submitted an interesting account of several au'hcntic 

 records of its occurrence, many of them abstracted from the ! 

 reports sent in to the United States Hydrographic Office. He , 

 showed that there was abundant trustworthy proof that afmo- ' 

 spheric electricity took on the form of globular lightning, but ' 

 conceded that the evidence was in details discordant. He hoped 

 that some day opportunity might be given to obtain a photograph j 

 of the phenomenon. 



Mr. Thomas Gray read a paper on the relative values of the 

 magnetic and the electiodynamic methods of measuring electric 

 currents in absolute measu es. | 



In the subsequent discussion on the paper, Prof. Nichols, of j 

 Cornell Univer.--ity, stated that he had compared the constants 

 of one of Sir William Thomson's balances as certified by him, 

 with the constants as determined by the large standard galvano- 

 meter in Cornell University, and found them to agree perfectly 

 through the entire range — a result with which he was greatly 

 pleased. 



Prof. H. S. Carhart, of the University of Michigan, read two 

 papers before this Section, one on magnetic leakage in dyn::mos, 

 the other on an improved standard cell with low temperature 

 coefficient. The following abstracts of these papers will show 

 their gist and purport : — 



The first paper gives an account of some experiments to 

 determine the ma^neiic leakage of two dynamos of the Mather 

 and Edison type respectively ; the former was a small machine 

 of*5oo watts capacity, built by a student in the laboratory, the 

 latter an Edison of 5000 watts capacity. 



In the small machine the leakage was due to some projecting 

 machine steel studs which carried the armature yokes. The 

 number of magnetic lines cut by the armature when the machine 

 was coupled as a self-exciting shunt dynamo was determined, 

 and then the steel studs were replaced by brass and the de- 

 terminations were repeated. The result was 190,000 and 252,000 

 lines of force respectively, or a gain due to removing the steel 

 studs of 32'6 percent. This was partly due, however, to increase 

 in the field when the leakage was stopped. Hence the machine 

 was independently excited by a storage battery, and the number 

 of lines determined in the two cases as before. The result was 

 164,600 and 184,100 lines cut by the armature with and without 



steel studf, a gain of li'2 per cent. This measured the" true 

 leakage due to the assigned cause. 



The leakage in the Edison machine was due to the iron' bed- 

 plate. A plank base was made, and the numl er ff magnetic 

 lines cut by the armature was determined first with the iron base 

 and then with the plank, the machine being self-exciting and 

 running at the same speed in the two cases. The numterof 

 lines was 1,510,000 and 1,605,000, a gain of 6*25 per cent, due 

 to removal of the cause of leakage. 



In his second paper, on an improved Clark standard cell with 

 low temperature coefficient, Prof. PI. S. Caihart said: — "The 

 objections to the Clark cell, as described by Lord Rayleigh, are 

 high and variable temperature coefficient, objectionable me- 

 chanical construction, and serious local chemical action. These 

 I have, at least in large measure, oveicome. The cell which I 

 now make is so constructed that the mercury cannot reach 

 the zinc during transportation. The temperature coefficient is 

 represented by the equation 



E^ - E[i - •ooo387(/ - 15) -f •oooooo5(/ - 15)-]. 



Per degree the coefficient is 



- '000386 -f •oooooi(/ - 15). 



At o°C. this coefficient becomes - '000401 ; at 25°, - 'C00376 ;. 

 and at the highest temperature observed, S2°"7, it is - '000348. 

 It will be observed that this coefficient diminishes with rise of 

 temperature, while that of Lord Rayleigh's cells increases. 

 In cells of the old form local action is due to zinc replacing 

 mercury when in contact with the solid mercury salt. The zinc 

 is thus amalgamated, and the amalgam creeps up. This I 

 prevent by keeping the zinc out of contact with the mercury 

 salt. The same device has the effect of raising the E.M.F. about 

 04 per cent. The polarization of these cells amounts to only one 

 ten-thousandth part in five minuies with an external resistance of 

 10,000 ohms. It reco^ ers in a few minutes. The different celli* 

 show great uniformity of E.M.F. All compaiisons are made by 

 Lord Rayleigh's method slightly modified. A charge or dif- 

 ference of E.M.F. of one ten-thousandlh part is very readily 

 detected." 



In this Section Messrs, W. A. Rogers and R. T. Woodward, 

 in a paper on mercurial thermometers, held the following 

 theses : — (i) The movements of a mercurial column are in all 

 cases by pulsation ; (2) these pulsations have a regular recurrence j 



(3) the period of recurrence is constant in the same thermometer ; 



(4) every pulsation has the same harmonic relation ; (5) the 

 amplitude of the curve which represerts the harmonic is in- 

 constant ; (6) as the period is constant and the time required for 

 the completion of the cycle variable, it follows that the danger 

 of error in random readings of the thermometer is greater for 

 slow than for rapid variations of temperature. 



Section of Biology. 



In the Biological Section, Mr. A. J. Cook read an interesting 

 paper on the alimentary apparatus of the bee, in which he em- 

 bodied the results of a long series of personally conducted 

 experiments. He differed, he said, widely in many particulars- 

 from the author of the article on the anatoiuy of the bee in the 

 " Encyclopaedia Britannica," notably as regards the conformation 

 of the tongue. This organ, Mr. Cook strongly maintained, was 

 a hollow cylinder furnished on the underside with a slit through- 

 out its entire length. He explained that suction could be per- 

 formed in three ways : (i) through the terminal aperture of the 

 internal tube, when the nectar could be reached only by the end' 

 of the tongue ; (2) through the slit opening into this tube, wheO' 

 the fluid to be sucked in was shallow and placed on a flat 

 surface ; (3) by the aid of the labial palpts when the fluid was 

 abundant. The author also expressed his incredulity as to the 

 possibility of a trustworthy analysis of honey, arguing that ii> 

 many cases more nectar was taken into the honey-stomach by the 

 bee than the glands had time or secretion sufficient to digest. 

 Especially, he thought, did this happen when the linden was in 

 bloom, when a single hive of bees would sometimes store up 15 

 pounds of honey in the day. The queen bees and the drones, he 

 held, were fed with digested food only, and to this fact he traced 

 the extraordinary fecundity of the queen bee. 



Prof. Burrill read a paper on fermentation of ensilage. He 

 stated that all fermentation of organic matter is now universally 

 admitted to be due to the action of micro organisms, and he 

 described the phenomena presented in recently stored gree» 

 fodder, used for cattle fcod. This material is now placed ii> 



