278 



♦ KNOWLEDGE ♦ 



[October 1, 1887. 



opposite in character, but as fatal in its effects, unless 

 corrected by devices not depending on chance — faith in 

 gambling systems, or what are sometimes called martin- 

 gales. 



These systems, some of which I have described eUewhere, 

 are singularly illusory. To study their working, not only 

 as given in detail by a believer, but as carried out again 

 and again in practice, they seem practically peifect. In 

 every case, the plan is guided by the gambler's actual loss 

 and gain ; when he has lost he wager-s somewhat more at the 

 nest turn than before his loss ; when he has gained he wagers 

 somewhat less, the actual diflerence depending on the par- 

 ticular system employed. But always after continuing the 

 system long enough, it results, unless there is an improbable 

 run of bad luck, that the losses are all cleared off, and a 

 certain sum, the same in every trial, is secured. It can be 

 shown, not only vaguely as by the believer in the system, 

 but by mathematical demonstiation, that the chances are 

 greatly against failure to secure this definite sum, in any 

 given trial of the system. Theoretically, indeed, if the 

 gambler's pocket is but long enough, and the limit put by 

 the bank on stakes is high enough, the process is certain to 

 end in the gambler pocketing his 10/. or 20/., or whatever 

 the amount he assigns at the outset for each trial, may be. 

 So he may go on winning twenty, thirty, forty, perhaps a 

 hundred times in succession, to the envy of all those who 

 are not acquainted with the plan on which the system is 

 worked. He may improve the system until he might get a 

 thousand successes for one failure. Thus faith grows in such 

 a system until the gambler assures himself that he has the 

 secret of securing unlimited wealth. Yet, all the while, 

 every step of his career is towards ruin. 



Tlie tiouble lies here : 



The surer the method of the system the smaller is the 

 sum made at such small risk as remains on each trial. This 

 circumstance is disguised by the fact that the risk ari.ses not 

 only in each trial regarded as a whole, but in the progress of 

 each trial. For instance, the trial having begun and gone 

 unfavourably for several steps — which is always likely 

 enough — the ri.sk that a fatal run of luck will occur before 

 the losses are wiped off is now very much greater than it 

 was at the beginning of that trial. Or if, after all has gone 

 well at first, a run of bad luck comes, then, though the 

 player is no worse off as to risk than at the outset, he has 

 lost all the benefit of the run of favourable luck, and is 

 situated as if starting on a fresh trial. The gambler blinds 

 himself to this, though every stake his system calls on him 

 to deposit should remind ham of it ; but in reality this is 

 where tlie danger chiefly lies. For the rest, it is easy to see 

 that the high probability there always is of winning at 

 any given trial affords no assurance against loss. If a man 

 A says to another, B, you shall draw again and again from 

 this urn, in which — as you can see and count — there are 

 one thousand balls all white except one, which is black, 

 replacing the drawn ball after each trial, and every time you 

 draw a white ball you shall pay me 1/., while every time 

 you draw the black one I will pay you 1,000/., A would be 

 most likely to have a long series of successes before he had 

 to pay 1,000/. Yet in the long run A would not be likely 

 to gain by such a venture, and he might lose very heavily. 

 If he agreed to take only ISs. from B at each drawing of a 

 white ball — which would conespond with the po.sition of a 

 gambler working out a system at a public gaming-table— he 

 would be sure to come off second best, if he only continued 

 long enough. 



And that is actually the way with all such systems. They 

 are pretty sure to work successfully many times before there 

 comes the run of luck on which they fail ; but that run is 

 quite sure to come long before the gambler has made as 



much by the system as will repay him for his time, trouble, 

 and worry ; and each failure will, on the average, sweep oft' 

 more than the previous successes have secured in the way of 

 gain. 



THE NATURALIST'S LABORATORY. 



CONTRIBUTION VIII. 



LABOEiTOEY EeQUISITES. 



|HERE are many things which add very 

 materially to the comfort of the student, 

 and convenience in the prosecution of his 

 studies, that have never been discussed 

 in any systematic treatises, but are never- 

 theless of the first importance, since they 

 usually draw the line between slovenly 

 and neat work in the laboratory. 



Preserving and Afacerathu/ Paiis. — It is well known that 

 the dissecting-rooms of most of our universities and colleges 

 are remarkable for then- pestiferous atmosphere. Where 

 human anatomy, or the structure of animals much over the 

 size of a rabbit or cat is taught, this undesirable state of 

 things can be minimised by careful previous treatment of 

 the "bodies'' and " parts " in question; but it cannot be 

 wholly subdued without enormous expenditure and unneces- 

 sary labour. In ordinary biological laboratories and in 

 private rooms however, anything of the kind above alluded 

 to ought not to be tolerated, especially as nothing much 

 larger than a cuttlefish or a cat need be dissected ; and most 

 of the large firms of potters, such as Doulton & Co., of 

 Lambeth, and others, manufacture glazed earthenware pans, 

 with suitable, closely- fitting covers, of various shapes and 

 sizes, for manifold uses in the household and kitchen — e.g., 

 ham pans, rinsing red pans, itc. The animal under operation 

 should be wrapped in a calico cloth well wetted with strong 

 methylated spii'it, and then immersed or partially submerged 

 in diluted spirit in the pan, during the intervals between 

 work. Where no suitable cover is provided with the pan a 

 piece of window glass cut approximately to its shape may 

 be used, and total exclusion of air, for even a protracted 

 period of time may further be secured by a liberal use of 

 hog's lard, after the fashion of a cement. Twenty such vats 

 in a private room cannot, under such circumstances, po'lute 

 its atmosphere, or prove otherwise offensive. Similar vats 

 without covers are decidedly preferable to wooden tubs for 

 macerating work, which, it need scarcely be remarked, 

 ought always to be conducted out of doors, or rather, in a 

 special outhouse or shed. 



Tlie Dissecting Dish, as its name implies, is useful for 

 animals of small size only, such as earthworms, snails, frogs, 

 (fee. Although an ordinary pie-dish can, and has, largely 

 been used for this purpose, it is unquestionably a very im- 

 perfect article. Let us take, for example, a frog : to learn 

 its anatomy thoroughly, several days of work should be 

 spent upon its dissection. The dish should be filled to the 

 depth of about an inch and a half with a suitable mixture 

 of paraffin wax and hog's lard melted together at a low tem- 

 perature, and jioured, whilst still fluid, but on the verge of 

 becoming solid, into the dish ; this will prevent any marked 

 after- shrinkage. The animal should next be fastened upon 

 the paraffin when solid, w-ith pins, and covered, or partially 

 covered with dilute spirit. After a day or two, when some 

 critical portion is about to be examined, the student often 

 finds, to his chagrin, that the liquid around his dissection 

 has insinuated itself between the sides of the dish and the 

 edges of the paraffin bed, by an almost imperceptible 

 shrinkage of the latter, sufficient, however, to render it so 



