that it is only in the rays, and some neighbouring genera, which have spiracles, that they are observed almost in the mouth. In fact it is to be presumed, that the liquid, in traversing them, would have deadened the sensibility of their surface by the rapidity of its motion, and the friction of its particles.

"Now are these peculiarities of structure, which I have mentioned, of such a kind as to lead us to abandon our first opinion, deduced from the knowledge of physics, that smells canot be perceived in water? or is not this supposed organ of smell in fishes better adapted to excite in them the sensation of tastes These questions I shall proceed to examine.

"Tastes and smells are nearly of the same nature: both sensations are produced by the physical and chemical qualities of bodies. We know, in fact, that very minute particles are continually separating from certain substances, which, without being decomposed, come to act immediately upon animals at that point of their surface alone, where they can manifest their presence. This phenomenon is effected by the in

tervention of a fluid medium, and a sort of contact.

"All the conditions necessary for the impression or sensation of taste are united therefore in the organ under examination, and the nature of the substances that may produce it. First, the organ is placed secure in a cavity: it opens and shuts at the will of the animal, it admits or rejects emanations at pleasure. Secondly, the sentient surface receives numerous and bulky nerves from the fifth pair; it is soft, moist, and mucous; and it presents a great surface in a large space. Thirdly, it appears in a certain degree to supply the place of the organ of taste, which cannot exist in the mouth of fishes from the very mechanism of their respiration.

"It seems to follow then from all these cirtumstances, that the organ of taste in fishes does not reside in the mouth: that the sensation of taste is probably imparted to them by the apparatus, which bas hitherto been considered as adapted to perceive the emanations of odorate bodies: and lastly, that no real smell can be perceived in water.

1

T

HOSE who have occasion to construct machines intended to be moved by men or animals, are sufficiently aware how im7portant it is to be acquainted with the quantity of power that can be attributed to either of them, in order to estimate with accuracy the effect

which it is proposed to obtain from the machine. It is well known, that the arrangement of the whole depends entirely on the ratio of the velocity of the motive force to the resistance. This was the reason that long ago induced experimen◄ talists to take the trouble of deter

mining the strength as well as the velocity exerted by men and animals, when they are made to move machinery; and the results they obtained, which have been commonly made use of in computing the effect of machines, are, that men exert from twenty-seven to thirty pounds, with a velocity of from one and a half to two feet per second; and that a horse has about seven times more strength than a man, with a velocity of from four to six feet per second.

"These are the data which we have been obliged to use whenever it became necessary to compute the effect of a machine moved by men or horses. It is evident that the force must be diminished when the

p=P (1 →

2

velocity is increased, and vice versa : but we are not yet certain of finding the ratio of the diminution or augmentation of this force to the velocity. Euler has given us two different formula to compute this ratio: but no one has hitherto attempted to verify by experiment which of them is to be preferred, although they differ very considerably from each other. If we put P for the absolute force which takes place when we simply consider equilibrium, C the absolute velocity which takes place when the man or animal moves freely, and without being overcome by the resistance, p the relative force, and c the corresponding velocity, we have by the first of these formulæ,

C

I am now about to detail, which of course would have been very expensive, had I not had some facilities which other persons may not possess.

"To make the experiments on human strength, I took promiscuously twenty men of different sizes and constitutions, whom I measured and weighed; the result of which is given in the following table:

Weight.

2

7

4 8

134

12

165

13

131

14

5

177

15

10

192

"To find the strength that each of these men might exert to raise a weight vertically, I made the following experiments :

"I took various weights, increasing by 10lbs. from 150lbs. up to 250lbs. All these weights were of lead, having circular and equal bases. To use them with success in the proposed experiments, I had at the same time a kind of bench made, in the middle of which was a hole of the same size as the base of my weights: this hole was shut by a circular cover, which effected this purpose when pressed against the bench, but at other times was kept at about the distance of a foot and a half above the bench, by means of a spring and some iron bars. To prevent the weight with which this cover was loaded during

150 160 6"

the experiment, from forcing down the cover lower than the level of the surface of the bench, I had several grooves made in the four iron bars, which sustained the cover at any height at which it might arrive by the pressure of the springs, as soon as the pressure of the weight ceased.

"After having laid the 150lbs, on the cover, and the other weights in succession, increasing by 10lbs. up to 250lbs. I made the following experiments with the men whose size and weight are given above, by making them lift up the weights as vertically as possible all at once, and by observing the height to which they were able to lift them. The following table gives the heights observed for the different. weights marked at the head of the table.

2 710

66

5597.

9 7

o"

I

2

1 3

4 I

2

0 2 I O

1

9

2 5 183

7 I

5 6 39

4 1

2 9 1

1 6

"This table proves to us, that the size of the men employed to raise the weights vertically, has consider able influence on the height to which they severally brought the same weight. We find also by this, that the height diminishes in a much more considerable ratio than the weight increases; and we may therefore conclude, that it is advantageous to employ large men when it becomes necessary to draw vertically from below upwards; and, on the contrary, it is more advantageous to employ men of con

siderable weight, when it is required to lift up loads by means of a pulley, about which a cord passes, which the workmen draw in a vertical direction, from above downwards. To find the absolute strength of these men in a horizontal direction, I took the following method:

"Having fixed over an open pit a brass pulley, extremely well made, of fifteen inches diameter, whose axis, made of well-polished steel, to diminish the friction, was threefourths of an inch in diameter; [ passed over this pulley a silk cord

worked

worked with care, to give it both the necessary strength and flexibibility. One of the ends of this cord carried a hook to hang a weight to it, which hung vertically in the pit, whilst the other end was held by one of the twenty men, who, in the first order of the following experiments, made it pass above his shoulders; instead of which, in the secondi, he simply held it by his hands.

"I had taken the precaution to construct this in such a manner, that the pulley might be raised or lowered at pleasure, in order to keep the end of the cord held by the man always in a horizontal direction, according as the man was tall or short, and exerted his strength in any given direction.

"I had made the necessary ar

"When the cord passed over the

rangements, so as to be able to load successively the basin of a balance which I attached to the hook at the end of the cord which descended into the pit, whilst the man who held the other end of the cord employed all his strength without advancing or retracting a single inch.

"The following table gives the weights placed in the basin when the workmen were obliged to give up, having no longer sufficient strength to sustain the pressure occasioned by the weight. To proceed with certainty, I increased the weight each time by five pounds, beginning from 60, and intervals of time, having always precisely a space of ten seconds between them. The result of these observations, repeated several days in succession, is contained in the following table: shoulders of the workmen

"It is necessary to mention, with regard to these experiments, that I took care to place, at certain distances, persons in whom I could place confidence, in order to observe whether these men marched uniformly and sufficiently quick without running.

"Having thus obtained, not only the absolute force, but the absolute velocity also, of several men, I took the following method to determine their relative force.

"I had made use of a machine composed of two large cylinders of very hard marble, which turned round a vertical cylinder of wood, and moved by a horse, which described in its march a circle of ten Rhinland feet. This machine appered to me the most proper to make the following experiments, which serve to determine the relative strength that the men had employed to move this machine, and which I use hereafter to determine which of Euler's two formula ought to be preferred.

"To obtain this relative force, 1 took here the same pulley which served me in the preceding experiments, by applying a cord to the vertical cylinder of wood, and attaching to the other end of the cord, which entered into an open pit, a sufficient weight to give successively to the machine different velocities.

"Having applied in this manner a weight of 215lbs. the machine acquired a motion which, after be1813.

ing reduced to an uniform motion, taking into account the acceleration of the weight of the friction, and of the stiffness of the cord, gave 2:41 feet velocity; and having applied in the same manner a weight of 220lbs. the resulting uniform motion gave a velocity of 2:47 feet. I only mention these two limits, because they serve as a comparison with what immediately follows. I began these experiments with a weight of 100lb. and increased it by five every time, from that number up to 400lbs.

"I made this machine move by the seven first of my workmen, placing them in such a way, that their direction remained almost always perpendicular to the arm on which was attached the cord which passed over their shoulders in an almost horizontal direction.

"Thus situated, they made 281 turns with this machine in two hours, which gave for their relative velocity c=2'45 feet per second. We have also the absolute force, or P, from these seven men by the above table = 730lbs. and their absolute velocity, or C = 5.30 feet.

"Therefore, by substituting these values in the first formula, we find the relative force p-250lbs. which agrees very well with what we have just found above.

"If instead of this first formula, the second be taken, it gives p= 153lbs. which is far too little.