512 



NATURE 



[February i6, 191 i 



instance, gives an occasional quivering motion to one or 

 both wings which is clearly perceptible to the unaided eye, 

 although propulsion and change of position relatively to 

 air currents seem to be accomplished by strokes of the 

 wings resembling sculling strokes. 



It is not the birds, but certain insects, which exhibit 

 quiverings of the wing imperceptible to the eye. The 

 hoverer-fly, Syrphus, for example, can remain in one spot 

 in the air while the wings are vibrating at such a rate as 

 to be invisible, and at the approach of danger, or at will, 

 it may suddenly by some movement, also invisible, transfer 

 itself to a distance of a yard or more, and there continue 

 the wing quiverings, which maintain the body almost 

 stationary. 



Is not motion in all flying and swimming things attained 

 by presenting the wings or fins at a suitable angle to the 

 air or water, while at the same time giving a pro- 

 pelling motion to the tail or dorsal fin and body, and also 

 by a sculling motion of the wings or side fins, in the case 

 of some insects and fishes, invisible to the human eye? 



Dei by, February 9. Edward D. Hearn. 



Demonstration of Peltier and Thomson Effects. 



The following method of demonstrating the Peltier and 

 Thomson effects may be of interest. In Fig. i the current 

 passes through an Sb-BJ-S6 bar, the points of contact 

 being amalgamated to reduce the resistance. Two coils 

 of No. 36 covered copper wire are wound on the bismuth 



A 



one near each junction, and by means of the leads A and B 

 are placed in the gaps of a metre bridge, and a balance 

 produced. On passing a current of i ampere through 

 the bars, one junction is heated and the other cooled, 

 which is indicated by a galvanometer deflection of about 

 40 mm. due to the change in resistance of the copper 

 coils. The direction indicates a heating where the current 

 flows from Sfe to Bt, and vice versa. 



Fig. 2 shows a similar arrangement for exhibiting the 

 Thomson effect. The bent iron rod is heated to red heat 

 at C, and the ends A and B dip into vessels of mercury. 



Fig. 2. 



thus ensuring a large temperature gradient. On passing 

 a current of 10 amperes in the direction ACB, AC is 

 warmed and CB cooled, showing that the Thomson 

 coefficient is negative. The part EF must be packed in 

 asbestos wool to prevent heating disturbances from out- 

 side. S. G. Starling. 

 Municipal Technical Institute, Romford Road, 

 West Ham, E., January 28. 



The Formation of Spheres of Liquids. 



In conducting Plateau's experiment for the formation 

 of_ spheres of liquid in a medium of equal density, it is 

 still customary to use oil of some kind in a mixture of 

 alcohol and water. The following method will be found 

 much simpler and more effective. A glass beaker about 

 10 cm. diameter and 15 cm. high is filled with water at 

 22° C. to two-thirds of its height. Bv means of a pipette, 

 NO. 2155, VOL. 85] 



100 c.c. of a solution of 30 grams of common salt in 

 I litre of water are discharged at the bottom of the 

 beaker, so as to form a lower layer slightly denser than 

 the water above. A large funnel furnished with a tap, 

 and having a stem i cm. or more in diameter, is now 

 placed centrally in the beaker so that the stem terminates 

 about 7 cm. from the bottom of the vessel. A quantity 

 of commercial orthotoluidine, at a temperature less than 

 22°, is poured into the funnel, and the tap turned so as 

 to allow the liquid to flow gradually into the water. A 

 sphere of orthotoluidine forms on the end of the stem, the 

 growth of which resembles that of a soap-bubble blown 

 from a pipe. 



It is quite easy in this way to make spheres 6 or 8 cm. 

 in diameter, and the red colour of the orthotoluidine 

 renders the procedure visible from a distance. The funnel 

 may be lifted out and the sphere left floating in the 

 water ; and on surrounding the beaker by a square glass, 

 vessel, also containing water at 22°, the true spherical 

 shape of the drop is seen. If the beaker be surrounded 

 by cold water at 15°, the sphere will elongate in its hori- 

 zontal diameter and sink, whereas if the surrounding 

 water be at 27° or more, a vertical elongation will take 

 place, and the sphere will rise and attach itself to the 

 surface of the water in the form of a hanging drop. This 

 behaviour is due to the fact that orthotoluidine and water 

 are equal in density at 22°, but owing to the former liquid 

 possessing a higher coefficient of expansion, it becomes 

 less dense than water above 22°, and more dense at a 

 lower temperature. 



It may be added that all the usual experiments with 

 liquid spheres can be carried out in the beaker, and the- 

 method of formation has the advantage that a sphere of 

 any desired size may be formed by closing the tap when 

 the requisite quantity of liquid has run out. In the course 

 of a general investigation of liquids which are lighter or 

 denser than water, according to temperature, the writer 

 has found several which may be made to produce spheres 

 at certain temperatures in the manner described, but has 

 found orthotoluidine to be best suited to the experiment. 



Chas. R. Darling. 



City and Guilds Technical College, Finsbury, E.C. 



Colliery Warnings. 



I. HAVE read the letters which have appeared on this 

 subject with considerable interest. We have two theories 

 before us. Both theories connect the presence of fire- 

 damp with changes of atmospheric pressure, but the one 

 considers a time of high pressure as being most likely to 

 cause an outrush of gas, whilst the other regards a falling 

 barometer as the period of greatest danger. It does not 

 seem at all reasonable to suppose that the atmospheric 

 pressure would compress the rock and force out the gas 

 as the Author of the Warnings suggests. Rather would 

 air enter the rock cavities in such circumstances. The 

 tendency for firedamp to escape during a falling barometer 

 would be greater than during a rising barometer, but the 

 evidence only shows a very slight connection to exist 

 between the rise or fall of the barometer and colliery 

 disasters. 



The firedamp generated in certain coal measures exists 

 in the rock,, apparently, under considerable pressures, and 

 its escape does not appear to be likely to be much affected 

 by atmospheric pressure changes. The Author of the 

 Warnings remarks : — " There was a time when no one f 

 guessed that the earth's surface was always on the | 

 move. ..." In colliery districts the earth's crust is i 

 always on the move, owing to the colliery workings them- 

 selves. This movement is not a bodily oscillation — it is 

 an actual rending of the strata for some distance below 

 as well as above the seam being worked. Is it not 

 likely that it is to the formation of fissures in the rock 

 in this way that the gas owes its liberation? Considerable 

 spaces may also be formed by the settling and creep in 

 front of a working face ; the firedamp would collect in 

 such spaces and be forced out by further settling. At any 

 rate, it seems clear that the escape of firedamp in quantity 

 is more likely to be the result of some local change rather 

 than to changes of atmospheric pressure. 



R. M. Deelev. 

 Inglewood, Longcroft Avenue, Harpenden, 

 February 3. 



