July 15, 1887.] 



SCIENCE. 



33 



hollow tube was distinctly visible, the centre being clear like glass, 

 while the outside was wrapped in a smoke-like mist. Even with the 

 naked eye we could distinctly see a spiral motion on the inside of 

 the tube, as if water were either ascending or descending, in which 

 direction it was impossible to tell. Beyond the waterspouts, be- 

 tween the cloud and the sea, a blue sky with sunlit cumulus 

 clouds was plainly seen. These tubes moved at the rate of more 

 than thirty miles an hour, judging from the time ordinarily taken by 

 steamers in crossing the same space. Estimates of the height of 

 the cloud are difficult to make, but at least half the tubes were 

 seen over the promotory of Capo delle Mele, which is about one 

 thousand feet high, and distant about live miles, as the crow flies, 

 from the point of observation. 



The phenomenon caused a great panic among the inhabitants, 

 owing to the prediction of Falb that there would be a violent earth- 

 quake on that day. 



There was subsequently a slight storm of hail and rain ; but 

 farther westward, on the coast, the damage done was considerable, 

 at San Remo secular olive trees being torn up by the roots and 

 whirled away. No waterspout is, however, known to have burst 

 on the land. Maurice Howard, 



Eugene Schuyler. 



Alassio, Riviera, Italy, May 26. 



HovT- to make Meteorological Observations at a Distance 

 above the Earth's Surface. 



The progress of meteorology in the beaten tracks of the usual 

 observations is very satisfactory ; but there are several new lines 

 of work, that can be and ought to be carried out, that receive 

 scarcely more than an occasional mention, or a regret that some- 

 body does not do something in the matter. The observation of the 

 conditions of the atmosphere above the earth's surface is perhaps 

 the most important of these questions. I know of no meteorologi- 

 cal data so much to be desired as that which is now obtained for 

 short, irregular intervals, by the occasional ascent of a balloon. 

 This, however, is a very expensive and risky method of observa- 

 tion, and has always been looked upon as a novelty rather than a 

 regular method. 



The few observations made in balloon-voyages, together with 

 those obtained by means of an occasional captive-balloon ascent, 

 are very valuable, and have been used over and over again in de- 

 termining constants. The great expense of even a captive balloon, 

 where the observer must go up. has prevented their general intro- 

 duction into meteorological work. 



It has often been proposed to send up self-registering instruments 

 in smaller captive balloons ; but, if this has been done, I have not 

 seen accounts of it. The lighter forms (metal thermometer and 

 aneroid barometer) could undoubtedly be used in this way ; but the 

 ordinary registering-apparatus is very delicate, and the swaying of 

 the balloon might disturb the adjustments ; besides, the original 

 cost of the apparatus is considerable, and, moreover, any damage 

 could not be easily repaired. 



In place of a balloon, the kite has been suggested, and E. Doug- 

 lass Archibald has made some interesting preliminary experiments 

 with this method. 



I have seen only the account of his experiments as given in the 

 Meteorologische Zeitschrift for 1885 (p. 47); but in this paper 

 there are references to Nature (Nov. 20, 1884) and Quarterly Joiir- 

 «a/ (January, 1883). 



Mr. Archibald flies two kites, the one to steady the other. He 

 carried on systematic observations with an anemometer (six inches 

 in diameter) for a year, and finally got results for a height of eleven 

 hundred feet above the ground. 



I saw this paper on Mr. Archibald's work a few days ago for the 

 first time ; but it interested me very much, as I had been consider- 

 ing the same problem. A year and a half ago I devised a form of 

 apparatus that would seem to promise good results ; but it was only 

 some months ago that I suggested the following detailed construc- 

 tion, which is given here for the benefit of any who might wish to 

 carry on any such experiments. 



The general form is seen in Fig. i, and consists of a balloon (5) 

 which carries a basket (4) suspended from beneath, and the b asket 

 carrying an anemometer (6) with a weight below it. 



The balloon is held captive by a three-strand (insulated) wire, 

 which is wound around a reel (2), and passes to a table (i), where 

 the battery and keys are mounted. 



The reel (2) must be firmly anchored, and the wires arranged so 

 there will be no danger from electric currents. A cloth ring, with 

 the length of rope (from the balloon) written on it, can be glued to 

 the rope at every hundred feet, so that the observer can see- 

 just how much rope is out ; and, by means of some instru- 

 ment for measuring vertical angles, the altitude of the balloon can 

 be measured and the height of the balloon computed. 



The apparatus as shown in Fig. 2 might also be sent up in a. 

 balloon held captive by an ordinary rope, if a small bichromate-of- 

 potash battery, with closed hard-rubber cells and a clockwork ta 

 break the circuit every five or ten minutes, is also included. The 

 whole apparatus might also be sent up on a kite, if one wished to 

 risk the instruments, which would be destroyed by the sudden 

 falling of the kite. 



The method here given allows the observer to control the time of 

 observation, and would seem best on that account. The register- 

 ing apparatus as shown in Fig. 2 is practically Professor Wild's sys- 

 tem, with some important differences, however. I was for a long 



time troubled about the means of moving the registration-paper 

 without clockwork, but Wild's method answers the purpose very 

 well. It must be borne in mind that only a very general description 

 of the apparatus is given here. 



The following apparatus is to be placed in the basket suspended 

 from the lower end of the balloon. The basket must be so ar- 

 ranged that the air will have free passage through it when the bal- 

 loon is ascending or descending. The balloon need only be large 

 enough to carry a few pounds (fifty) to the height of half a mile : 

 it is impossible to foretell just how much the whole apparatus would 

 weigh. A hair hygrometer could also be added to the instruments, 

 but has not been put in the accompanying sketch. 



The careening of the balloon would have no effect on the work- 

 ing of the apparatus as shown here, because nothing of the regis- 

 tration arrangement is free to move except the pointers. In making 

 an apparatus, the best arrangement would be somewhat different 

 from the sketch given here. 



The main advantage of this apparatus is the cheapness with 

 which an ordinary aneroid barometer and metal thermometer could 

 be applied to the purpose. 



In Fig. -z, R\%3. stand on which the long lever D rests, and turns 

 in the vertical. N and ISP are two electro-magnets which attract D. 

 When the key i joins 2-3, then N acts, and draws D down on the 

 right. When the key i joins 2-4, then N' acts, and pushes D up 

 on the right. Self-registering paper is coiled on the wheel c, and 

 one end of the paper passes between the rollers F and G, and then 



