nitrate of silver, an insoluble silver-salt was prepared in the usual manner for preparing these compounds.

Of this silver salt 0.4795 grms. was ignited, until all organic matter was destroyed, 0.1360 grms. =28.36 per cent. of pure metallic silver, remained behind.

Varrentrapp obtained from pure margarate of silver, 28.20 per cent. of silver, which agrees very well with the result which I have obtained. By a combustion with chromate of lead, 0.3241 grms. substance yielded Carbonic acid=0.8955 grms., and water 0.361 grms., which makes the composition of the substance: Carbon 75.35 per cent.; hydrogen 12.37 per cent. ; oxygen 12.28 per cent.

Varrentrapp's and Sacc's analyses vary between 75.30 and 75.85 per cent. carbon, and 12.22 and 12.69 per cent. hydrogen. These analyses prove that the solid acid in the fat operated upon, is identical with margaric acid. The fat could not contain any stearic acid, because if so, this would have accompanied the margaric acid throughout the process of its preparation, and would have raised its point of fusion above 60° C.

The soap which the original fat forms with soda, has a yellow tinge, which can be removed mostly when the soap is separated by salt out of a strongly alkaline solution. The alkaline mother-liquors are then of a deep brown color, and on super-saturation with an acid, deposit a small quantity of a brown sticky precipitate, which is soluble in alkali or in alcohol, and seems to consist of fat surcharged with coloring matter.

As the result of this investigation, it appears that the fat under examination consists of margarate and oleate of glycirin and a small quantity of the volatile acids of butter; it contains no stearic acid and shows great analogy with the fat of geese.

ON AN INSTRUMENT FOR DETERMINING THE VARIATION AND DURATION OF WINDS. BY PROF. JAMES H. COFFIN.

THE Report of the Regents of the University of the State of New York, for the year 1839, contains a description of an instrument that I formerly used to measure the duration of winds from the several points of compass. A vane was attached to the top of a perpendicular shaft, and at the bottom of the shaft a funnel-shaped tube was fastened pointing obliquely downward, so that as the shaft revolved by the action of the wind on the vane, the smaller orifice of the tube would describe a horizontal circle. A small stream of sand, gauged

in the same manner as in an hour glass, was constantly running into the tube at the upper orifice, and thence descending to the lower orifice, was distributed into different boxes, (thirty-two in number,) according to the direction of the wind. The quantity of sand collected in each box, showed precisely the length of time that the wind had blown from the corresponding point of compass.

The foregoing instrument, which I used about a year and a half, answered a very good purpose, but there were two defects in it. 1st. While it recorded the duration of the several winds with great minuteness, it failed to inform me at what hour and minute they occurred, a point of considerable importance in connection with the study of storms, and several other meteorological phenomena. 2d. It required attention twice a day to replace the sand.

With a view to remedy these defects I modified the instrument, by substituting for the stream of sand a row of minute cards, arranged at regular intervals upon a movable band or apron, each card having printed upon its face the number of the day and hour upon which it made the record. The motion of the band or apron was regulated by a clock, in the same manner as in the animometer of Osler.

As thus modified it seems to me to possess the following advantages over that instrument: 1st. It operates with more certainty, and as cards cannot fail to make their records at the proper time, whereas a pencil is liable to get out of order. 2nd. Its records are more definite, dividing the winds into thirty-two distinct classes, or a greater number if desired. 3d. It requires attention less frequently. There is no difficulty in making it keep an hourly record for months together without care. 4th. It not only shows the direction of the wind on any given day and hour, as is done by Osler's, but it also prepares its own abstracts, by collecting the records for each point of compass into a separate box, thus saving much labor in reducing them.

By the motion of the apron the cards are carried forward, so that at the end of each hour one arrives at the roller E, (see Fig.) and falls off into some one of the boxes below, the particular box being determined by the direction of the wind at the time. As each card shows upon its face the day and hour upon which it fell, the record is complete. For example, suppose that on examining the boxes at the end of the month of July, the cards which I enclose were found in the box labelled "South." It would show that there was a south wind on the 3d day, 18th and 20th hours; on the 18th day, 12th hour; on the 22d day, 24th hour, &c.

A, is the Vane.

B, is the Perpendicular Shaft.

D, is a small weight to relieve the Clock.

N, NE, E, &c., are paper boxes placed upon the circular plate, to receive the cards, as they fall from the apron at

In an instrument of this kind that I set up at an observatory on the summit of Saddle Mountain, near Williams College, and which was designed to run four months without care, during the colder part of the year, when the mountain is inaccessible, the apron was necessarily too long to pass around rollers in the manner represented in the figure, and it was therefore wound upon drums. But when the records can be transcribed as often as once a week, the form given in the figure is preferable.

I may remark farther, that the force of the wind may be registered in the same way, by means of an additional row of cards placed upon the apron.

Seventh Day, August 21, 1849.

SECTION OF NATURAL HISTORY, ZOÖLOGY, &c.

ON THE STRUCTURE AND HOMOLOGIES OF RADIATED ANIMALS, WITH REFERENCE TO THE SYSTEMATIC POSITION OF THE HYDROID POLYPI. BY PROF. LOUIS AGASSIZ.

THE importance of the study of homologies has been fully exemplified by the results obtained from a thorough comparison of the different classes of vertebrated animals. A deep insight into their structure has been gained since we knew that their different parts, as fins, wings, and legs, are only modifications of the same organs, undergoing different developments in the different classes. The homologies of the skeleton have been particularly investigated, and those familiar with these researches know that they have led to a thorough understanding of the typical peculiarities of this system of hard parts.

Less extensive, though not less valuable investigations have been made upon the muscular as well as upon the nervous system, with equally valuable results.

Besides the more accurate knowledge of organs thus obtained, another important conclusion has been derived from these studies. It has been shown that the functions of certain parts cannot altogether be relied upon in order to ascertain their true nature. For instance, it was a mistake to identify the lungs and gills among Vertebrata, and to compare the aerial respiratory organs of mammalia, birds, and reptiles, with the gills of fishes, since there are reptiles in

which we find both kinds of apparatus, existing simultaneously throughout life, and performing at the same time the same functions. This shows that in the investigation of the structure of animals, we should distinguish between homologies and analogies, just as well as we distinguish between affinities and analogies, where we investigate the natural relations between animals.

The whales have no affinity with fishes beyond the general one which exists between all Vertebrata, but they are analogous to the fishes in their general form, and by the fact that they inhabit the same element.

The bats are analogous to birds in being provided with wings to fly through the air, but they do not belong to the same class; their nearest relation is with the mammalia.

In the same manner should we distinguish between the different kinds of apparatus which perform various functions, as I have already pointed out, in mentioning the gills and lungs. But at the same time, it is important to ascertain the structural relation, or the homology which exists between kinds of apparatus performing different functions. As for instance, the legs of quadrupeds, the pectoral and ventral fins of fishes, and the wings and legs of birds, which are truly homologous, while their lungs and gills are only analogous to each other.

This first general result will presently lead to another very important investigation. It is a matter of great interest to naturalists, to ascertain how far the parts which perform similar functions in the different great groups of the animal kingdom, are homologous to each other, and how far they are simply analogous. Is it true, for instance, that the legs of insects, crabs, lobsters, and worms, correspond truly to the parts which we designate by the same names among vertebrates? Are the wings of insects identical in structure with the wings of birds? Or, are these parts only analogous? Can the respiratory apparatus in Articulata, their gills and tracheæ, their various air sacks, be compared to the gills and lungs of vertebrates? Is the heart of Crustacea the dorsal vessel of insects? Are the vesicular sacks of worms really hearts in the same sense as we distinguish a heart in vertebrates?

Again, are the gills and lungs of Mollusca, their various apparatus of locomotion, identical with the similar parts in Vertebrata and Articulata ? And can we trace correctly such a comparison among

Radiata?

This is the question which I propose to answer in this investigation