attraction or repulsion; advantage should be taken of this circumstance to direct the current in the manner most convenient.

It may be further added, that when the needle has deviated a certain number of degrees, and it is wished to observe minute changes of temperature, and consequently, by their thermo-electrical effects, the current should then be so directed as to lead the needle towards Zero, because it acts with so much more power, when its direction is, less oblique in relation to that of the needle.

The temperature of the mouth may, serve as a term of comparison in the absence of a better, but as frequent variations are always to be feared, depending upon the manner in which the soldered-junction is placed, this mode ought to be rejected whenever delicate examinations are entered upon. However, there is a means of verification, even in this case, which ought not to be omitted to be stated. The arrangement should be inverted, that is to say, that the soldered-junction of the second probe, or that which has been in contact with the mouth, should be placed in the part where the temperature is required. If the results are the same, their correctness is certain; if contrary, the cause of the difference must be sought after, and the experiment continued until an absolute equality is attained.

It has been found, after many attempts, that the apparatus delineated in fig. 7, has the advantage of conveniently affording a constant temperature.

A small wooden vessel (A), lined with sheet-lead, and having a wooden cover (H), is prepared. The cover has an aperture, by which a thermometer is introduced, and another through which is passed the probe, whose soldered-junction (b) is intended to be maintained at a constant temperature. This will be about 90° when mammalia are the subjects of examination.

Water at 122°, on being first poured into the vessel, suffers a depression of temperature, which can be permitted to any degree wished. The vessel is then to be placed in a receiver (B), whose height is rather greater, and into which water, heated to 104°, is to be poured. This temperature has been found to be of that height that the thermometer in the vessel (A) is not sensibly depressed when this falls a degree. The receiver is intended to impede the loss of heat in the vessel, and it is desirable that its arrangement should be such that the water in the receiver shall be maintained within a degree or so at the same temperature. Two means may each be employed for this purpose; one is to renew the heat in the receiver from time to time, by pouring in warmer water by means of a tube, removing at the same time an equal quantity of the water which has been cooled. But this operation is troublesome, and it may be substituted by the following, in which an apparatus regulates the introduction of the warming water, and the withdrawal of that which has been cooled.

On a stand (T) at a small distance from the receiver, is placed a vase (c) of tin plate; from this a copper tube (t) furnished with a cock (r) is carried into the receiver, and down to near its bottom. Another cock (R) is attached to the lower part of the receiver. Water is to be maintained in the vase (c) at a temperature of about 160°, by means of a

264

INTERNAL TEMPERATURE OF ANIMALS AND VEGETABLES,

lamp (L), its cock (r) is opened, and the hot water enters the receiver and is carried to its bottom; as this warmer water ascends in the receiver, it rewarms the whole apparatus; the receiver-cock (R) is now opened, and a proper quantity drawn off. With a little practice, and frequent observation of the thermometer, the constant temperature desired may be maintained in the vessel (A). One of the probes (abc) is introduced into the vessel, and another (a' a') is inserted into any muscle whatever, and their ends connected and brought into communication with a multiplier (G). It will be necessary to construct previously a table of temperatures. Suppose it is proposed to operate upon the genus Mammalia, and that the temperature of one of the soldered-junctions is to be maintained at 97°, and the other plunged into a vessel of water, whose temperature can be varied from 86° to 112°. The deviation corresponding to each change of temperature must be noted, and when all the observations are tabularly arranged, the temperature corresponding to each deviation may be obtained on inspection.

With these means we may proceed to examine the temperature of the animal world.

[To be continued.]

REVIEW.

I.

A Treatise on the principal Mathematical Instruments employed in
Surveying, Levelling, and Astronomy: explaining their Construction,
Adjustments, and Use, with Appendix and Tables. By F. W. SIMMS.
Second Edition, improved and enlarged. 8vo., 128 pp., Cuts. London,
Weale.

Ir would be difficult to find anywhere so much valuable information, conveyed in so masterly a manner, as is done in the work before us. Mr. F. Simms is peculiarly well qualified for his task, from his connexion with the partner of the late Mr. Troughton, who ranked as the first scientific instrument-maker we have yet possessed,-from having been several years an assistant in our national observatory, and-from having formerly been engaged in the Ordnance Survey of the kingdom. Mr. F. S. has thus enjoyed unusual opportunities of studying all the instruments he describes in their most varied and improved forms; of using the astronomical ones under the superintendence of Mr. Pond, one of the best observers of his time; and of employing the geodæsical ones under the most eminent masters of the art of surveying. To such extraordinary means of accurate acquaintance with his subject, he adds a power of conveying his information in an elegant, because simple and unaffected, manner; and he has, consequently, produced a work invaluable to the professional man, and interesting to every reader who possesses a taste for practical science.

The title gives but an inadequate idea of the contents of the book; for, in addition to a clear description of all the principal instruments used in surveying, and in astronomical observations, with an explanation of their various adjustments, Mr. F. Simms has omitted little which has any reference to the subjects of surveying, levelling, and observing; he has given instructions for keeping the field-books, registering and reducing observations, with the formula for the latter, these he has translated, as it were, into practical rules, for the sake of such learners as may not possess much mathematical knowledge; he has entered into details as to the mode of laying down, or plotting, a survey, and of copying or reducing plans, maps, &c., with a great variety of subsidiary matter; and has furnished a small collection of valuable tables relating to these subjects. In short, Mr. F. S. is one of those authors who, really anxious to instruct their readers, spare no pains which can facilitate their progress.

To enable our readers to judge for themselves of the merits we have been commending, we select the following passage, describing the corrections to be made for errors in the position of the portable transit instru

ment:

VOL. I.

U

4

"From the description which has been given of the method of bringing a transit-instrument into a state of perfect adjustment, it might be inferred that it is essential it should be strictly so, to obtain accurate results from the use of it. It is certainly desirable that the adjustments should be examined and rectified as often as possible, as doing so ultimately saves the labour of computing the corrections to be applied to each observation, on account of the errors in the position of the instrument. But in some established observatories, where large instruments are employed, it is not attempted to put them in perfect adjustment, but the amount of the various derangements is ascertained from time to time, and the observations corrected accordingly. The adoption of this method, with so small an instrument as the one which we have been describing, where the adjustments are easily examined and corrected, will give indeed more accurate results, but, on account of the greater trouble, is not perhaps to be generally recommended; we shall, nevertheless, introduce in this place, an account of the method of computing these corrections, that persons possessing transit-instruments may adopt which method they think proper.

"The first correction is for the deviation of the line of collimation: the amount of the error may be determined by a micrometer attached to the eyeend of the telescope, by which, when the telescope is directed towards any distant object, the angular distance of that object from the central wire is measured in revolutions and parts of the micrometer-screw. The instrument is then reversed, and the distance of the same object from the central wire again measured, when half the difference of the measures is the error in collimation : and the angular value of a revolution of the screw being known, the corresponding value of the error is likewise known. The correction on account of this error to be applied to the time of each observation may be computed from the following formula.

c the error of collimation + if the deviation is toward the east.
T= (as before) the polar distance of the star.

"Hence we have in words this rule: To the log. of the deviation in collimation, add the log. co-secant of the polar distance of the star, and the arithmetical complement of the log. of 15: the sum will be the log. of the correction in time required.

"The next correction to be considered, is that arising from a want of horizontality in the axis. The spirit-level, which we described as striding across the instrument and resting on the pivots, determines the amount of the inclination of the axis, and also, as we have seen, enables the observer to correct it. Above the glass tube, and parallel to its length, is placed a fine graduated scale, the reading of which points out the number of seconds in arc that the pivots deviate from the true level, shown by the air-bubble receding from the centre towards that pivot which is the highest; but as it is necessary, when correcting for the adjustment, to remove half the error, by giving motion to the little screw on the level itself, so, for the same reason, in finding the measurement of the error, it is necessary to reverse the level on the axis, and read the scale at each extremity of the air-bubble in both its positions; that is, with the same end of the level on both the east and west pivots alternately,and half the difference of the means of the two readings will be correctly the amount of deviation. This may be illustrated by the following example, in which the divisions on the scale represent seconds.