March 30, 1888.] 



SCIENCE. 



153 



many more experiments are needed, and that they should be of 

 such a character that each effect can be ascribed to its proper cause, 

 ■and that causes and effects shall not be treated collectively, as at 

 present. 



On Probabilities. 



A year ago, or more, Mr. M. H. Doolittle presented a paper to 

 the Mathematical Section of the Philosophical Society, on the doc- 

 trine of probabilities. It gave rise to an interesting discussion at 

 the time, which led him, at the last meeting of the section, to return 

 to the consideration of the subject. Referring to an important 

 change of opinion by John Stuart Mill, as shown in the eighth 

 edition of his ' System of Logic,' and set forth in the introductory 

 paragraphs of the chapter on ' The Calculation of Chances,' Mr. 

 Doolittle showed that the two antagonistic schools started with two 

 different definitions of the doctrine of chances, — one, to which he 

 belongs, accepting the latest definition by Mill, which he adopts 

 from Laplace ; and the other, that given by Mill in the first edition 

 -of his ' Logic' 



"Probability," says Laplace, " has reference partly to our igno- 

 rance, partly to our knowledge. We know that among three or 

 more events, one, and only one, must happen ; but there is nothing 

 leading us to believe that any one of them will happen rather than 

 the other. In this state of indecision it is impossible for us to pro- 

 nounce with certainty on their occurrence. It is, however, probable 

 that any one of these events, selected at pleasure, will not take 

 place ; because we perceive several cases, all equally possible, 

 which exclude its occurrence, and only one which favors it." 



" To a calculation of chances, then," says Mill, " according to 

 Laplace, two things are necessary : we know that of several events 

 some one will certainly happen, and no more than one ; and we 

 must not know, or have any reason to expect, that it will be one of 

 these events rather than another." Mr. Mill then expounds the 

 doctrine formerly held by himself, to the effect that these are not 

 the only requisites, and that Laplace has overlooked, in the general 

 theoretical statement, a necessary part of foundation of the doctrine 

 ■of chances, — the knowledge that one or the other of the events 

 must happen, but the possession of no grounds for conjecturing 

 which. "We must remember," explains Mill, "that the prob- 

 ability of an event is not a quality of the event itself, but a mere 

 name for the degree of ground which we or some one else have for 

 expecting it." 



Having read these passages, Mr. Doolittle took up briefly the dis- 

 cussion of the doctrine of philosophical necessitj', and referred to 

 Edwards on ' The Freedom of the Will ' as exceedingly able in the 

 presentation of this doctrine, and one of the first, if not the very 

 first, American book that became famous throughout the world. 

 On the other side, he quoted from Adam Clarke's ' Commentary 

 on the Bible ' as one of the ablest opponents of philosophical neces- 

 sity. Dr. Clarke's argument is, that, since there are events in the 

 future which are uncertain, it is impossible for them to be known 

 as certain, so that divine foreknowledge is only a knowledge of 

 probabilities, and does not include the certain knowledge of uncer- 

 tain things. Mr. Doolittle then asked his audience whether, what- 

 ever they might think of Adam Clarke's Deity, any one would claim 

 to be a Deity of that sort himself, and argued, that, in any case, it 

 is proper for us to base our theory of probability on human intel- 

 lectual conditions, and not on divine intellectual conditions. He 

 then said that the doctrine of probability is not peculiar in this re- 

 spect. Metaphysicians say that all our knowledge is based upon 

 our states of consciousness. We know only our states of con- 

 sciousness, and although we cannot say that any probabilities exist 

 in the nature of things, still we may presume that probabilities hav- 

 ing a scientific basis, have in some manner their counterparts in the 

 external world, just as we presume that other states of conscious- 

 ness have their counterparts in the external world. 



With regard to such probabilities, Mr. Doolittle said Mill was 

 right in his first edition. But there still are probabilities of less 

 scientific character that may nevertheless be made the subject of 

 mathematical computation. 



This paper was discussed for an hour by leading members of the 

 section. Professor Harkness of the Naval Observatory accepted 

 the definition of probabilities given by Mill in his first edition, as did 

 also several other gentlemen connected with that institution. The 



gentlemen connected with the Coast Survey, on the other hand, 

 generally accepted Mill's latest definition adopted from Laplace. 



Dynamite Guns. 



Among the appendixes to the ' Annual Report of the Chief of 

 Ordnance,' soon to be published, is one prepared by Maj. George 

 W. McKee, on ' The Present Status of Dynamite as an Explosive 

 for Shells.' Prefacing it with a brief history of the discovery and 

 use of nitro-glycerine, he says, — 



" The Nobel's explosive gelatine, or blasting dynamite, has been 

 used in this country by United States officers to the entire dem- 

 onstration of the fact that this high explosive, contained in a shell 

 as a bursting charge, might be fired from a gun. The ordinary 

 blasting dynamite made by the company (some of it experimentally 

 modified with about 3 per cent of camphor) was used, and enough 

 shells were thrown from the bores of the old mutilated guns used 

 in the experiment to demonstrate the fact that dynamite could be 

 projected in shells from an 8-inch rifle gun with a 40-pound charge 

 of powder. The great chemist Nobel never, perhaps, thought of 

 applying his invention to this delicate test ; but his powerful and 

 wonderful gelatine, made only to be detonated in mines and the 

 like, stood in several instances the tremendous initial shock of the 

 gunpowder, and, by the aid of the rectangular diaphragms devised 

 by Captain Whipple of the Ordnance Department, stood, what is 

 thought to be equally dangerous, the heat developed by the angular 

 velocity. If the gelatine had been especially undertaken by these 

 chemists for a military and not. an industrial agent, and enough 

 time and means had been at hand to perfect the diaphragm, 

 it is believed all of the shells would then have become, as 

 they will be in future, high-explosive batteries, projected with 

 as much safety as though they fiad been charged with black 

 gunpowder." 



Major McKee, in reviewing various experiments that have been 

 conducted under the direction of the Ordnance Department, speaks 

 of them as follows. Of the method exhibited by Mr. Snyder, 

 he says, " He did fairly well with some of his firing at the Hook 

 and on the Potomac, near Washington, D.C., and, as he is a man 

 of inventive talent and an American, no one wishes him more suc- 

 cess in his future experiments with dynamite than the men who 

 were delegated by the government to supervise and report upon 

 those he originally undertook." In the experiments with shells 

 loaded with dynamite, conducted by Brevet Brig.-Gen. John C. 

 Kelton, at Point Lobos, near San Francisco, Cal., in March, 1885, 

 no specially camphorated or otherwise prepared explosive was used, 

 but the shells were charged with the crude, blasting, industrial 

 dynamite. Three rounds were fired from a 3-inch wrought-iron 

 rifled gun, — shells with two hundred grams of dynamite, and a 

 variable charge of projection. The target was a large rock at 157 

 yards distance. In the first two rounds the shell burst into innum- 

 erable pieces on striking the rock, but in the third it burst within 

 the piece. Colonel Kelton considered this experiment as very satis- 

 factory, since it demonstrated the possibility of employing dynamite 

 in shells, as well as the great strength of this great explosive ; and 

 he estimates that for the effective use of these artifices, which, ac- 

 cording to him, is to destroy ships, one-half the length of the pro- 

 jectile is the penetration needed, requiring o.ooi of a second, and he 

 expects it will be successful. 



After describing some experiments at Sandy Hook in 18S3, 

 Major McKee sums up the results as follows : — • 



" As detailed in the records, three shells were fired with fulmi- 

 nate-of-mercury fuzes. The fulminate was too sensitive to stand 

 the shock, and it was found afterwards that the gelatine needed no 

 detonator. 



" Although the tests made were very few, it would nevertheless 

 appear from them — 



" (l) That the shells explode after clearing the muzzle, and there- 

 fore the detonation of the gelatine is due to some cause other than 

 the shock of discharge, very possibly the heat generated by angular 

 velocity. 



" (2) This is corroborated by the fact that one shell passed 

 through a 2-inch board target without explosion. 



" (3) The gelatine used in these tests, not being camphorated, 

 renders it highly probable that a certain percentage of camphor 



