May 12, 19 »o] 



NATURE 



1 1 



I 



fiving a kite with non-rigid edges at Pyrton Hill, and we 

 have been fortunate in breaking only one kite in landing 

 it, and in not failing once since last October to bring back 

 rhe kite to the starting point without accident ; also Mr. 

 Cody's kite, which has non-rigid edges, will certainly fly 

 in a strong wind. 



Mr. Fergusson states that a kite of 7 square metres 

 surface will exert a pull of 35 kilograms in a wind of 

 10 metres per second. In English units this is equivalent 

 to I lb. per sq. foot of sail area in a wind of 22^ miles 

 per hour. If the whole area were exposed normally to the 

 wind, the pressure or pull would be 53 kilograms, and 

 hence, remembering that the back sails are partially 

 sheltered by the front, and that the angle of incidence 

 is only about 15 degrees, 35 km. seems a very high value. 

 The pull of a diamond-shaped box kite of 7 metres sail 

 area in the same wind is certainly below 15 km. It 

 would be interesting if Mr. Fergusson would tell us how 

 the wind at the kite is measured at Blue Hill. I do not 

 think any anemometer placed in the kite can be trust- 

 worthy — one might as well place one close to the roof of 

 a house amongst a set of chimney stacks — and if an 

 anemometer is placed on the wire there is the difficulty of 

 avoiding oscillation and of correct orientation. 



I am glad to be able to state that we do not now officially 

 publish in England values of wind velocity based on the 

 factor 3 for the Kew pattern Robinson cups, but on the 

 factor 2-2. This reduces what would have been called 100 

 to 73 ; but the values quoted by me in my letter of 

 March 17 are' entirely inde{)endent of the Robinson anemo- 

 meter. For reasons fully given in a recent publication of 

 the Meteorological Office (M.O. 202) those values are 

 doubtful, but the evidence is in favour of their being too 

 low rather than too high. 



I should like to take this opportunity of replying to 

 Mr. Gold's criticism of the method of measuring wind 

 velocities on a kite in England, namely, by measuring 

 the tension of a piece of cotton carrying a light sphere 

 at the far end, away from, and out of the influence of, the 

 kite (Nature, April 21). It is true that the surface of 

 the cotton exposed to the wind is comparable with that 

 of the sphere, but the force is for all practical purposes a 

 normal- one, since the tSngential component is admitted by 

 all to be very small. It follows that the tension of the 

 cotton, neglecting its weight, is the same throughout, just 

 as in the case of the string stretched on ^ smooth cu-ve 

 given in text-books on mechanics, and hence the tension 

 measured is the tension of the cotton where it is tied to 

 the sphere, and is injjependent of the length. 



W. H. Dines. 



A Difference in the Photoelecttic Fffect caused by 

 Incident and Divergent Light. 



Recent investigations have shown that the ionisation 

 produced by the secondary rays arising from a thin metal 

 plate traversed normally by a primary beam of Rontgen 

 or 7 rays is greater on the emergent than on the incident 

 side. The present experiments were made to see if a 

 similar effect could be detected with ultra-violet light. 



Thin films of platinum were prepared by sputtering from 

 a platinum kathode on to quartz plates i mm. thick. 

 These could be mounted in the centre of two similar brass 

 cylinders so that their planes were perpendicular to the 

 axes of the cylinders. A narrow beam of ultra-violet light 

 from an arc passed down the axis of the two cylinders 

 normally to the plates. The saturation current from the 

 illuminated plates to the cylinders could be measured. 

 The plates could also be turned so that the film side was 

 either away from (position A) or towards the light (posi- 

 tion B). 



In every experiment two similar plates were used ; one 

 was used as a standard to determine the strength of the 

 ultra-violet light, and its position, whether A or B. was 

 unchanged. The other plate was compared with this for 

 each of the two positions alternately. By referring each 

 measurement to the standard plate, the otherwise trouble- 

 some variations of the arc were rendered harmless. Unless 

 the films were very thick it was always found that posi- 

 tion A gave rise to a relatively greater photoelectric current 



NO. 21 15, VOL. 83] 



than position B, although it was penalised by having to 

 pass through the thickness of the quartz plate. 



When no allowance is made for the absorption by the 

 quartz, a very thin film gives 12 per cent, more photo- 

 electric current for the emergent than for the incident 

 light. When the absorption of the quartz is allowed for 

 the difference is increased to ib per cent. 



These results have been confirmed by reversing the direc- 

 tion of the light without altering the position of the plates, 

 and other experiments have been made to ensure that they 

 do not arise from scattered light or other defects in the 

 apparatus. The ratio of the emergent to the incident 

 effect has been determined for a series of films of varying 

 thickness. 



This investigation was suggested to me by Prof. O. W. 

 Richardson, and the experiments have been carried out 

 under his direction. 



Otto Stuiilmaxx. Jln. 



Palmer Laboratory, Princeton. N.J., .April 26. 



A Link in the Evolation of the Bees. 



The ligula or " tongue "' of the bees presents two main 

 types, one broad, obtuse, and often emarginate, the other 

 pointed, acute, frequently much elongated. The obtuse- 

 tongued bees have been considered to be the more primitive, 

 and there is no doubt that the most advanced tjpes are 

 long-tongued. The difference between the two groups has 

 seemed so important that at one time (Trans. American 

 Entomological Soc, xxix.. p. 185) I entertained the idea 

 that they had no common bee-ancestry, but were derived 

 from different groups of wasps. 



Frederick Smith, in 1853, described a new genus of bees 

 from Australia under the name Meroglossa. This was 

 based on a male from Port Essington, which had many 

 of the characters of the obtuse-tongued Prosopis, but had 

 a pointed, dagger-like tongue. Ashmead, in 1899. placed 

 it in the same group as Prosopis, in spite of the tongue ; 

 in 1905 (Trans. Anier. Ent. Soc, .xxxi., p. 318) I gave an 

 account of Smith's type, remarking that it was *' not 

 unlike some Prosopis." In 1905 I described a number of 

 Australian species supposed to belong to Prosopis, but re- 

 marked of one of them (P. turneriana) that the mouth- 

 parts did not seem to agree with the genus. I had at 

 that time no material for dissection, but Dr. R. C. L. 

 Perkins had such material, and discovered that several 

 had acute tongues. In Proc. Hawaiian Entom. Soc., 

 October, 1908, he founded the genus Palaeorhiza for my 

 P. perviridis, with the following interesting remarks : — 



" Palaeorhiza is evidently represented by many sjjecies in 

 Australia. Several have been described as belonging to 

 the genus Prosopis, in spite of the fact that the most 

 superficial e.xamination shows that these insects have an 

 acute lanceolate tongue. Hitherto no connecting-link 

 between the blunt-tongued and acute-tongued bees has been 

 recorded, but in Paljeorhiza we have a form which, except' 

 for the structure of the tongue, would be assigned to the 

 section of Obtusilingues. It will therefore be obvious that 

 this section and the Acutilingues can no longer be main- 

 tained as of great importance, since Palaeorhiza must 

 always be associated with Prosopis, as the male genital 

 characters, and all other ones, save the lingual, clearly 

 s'now." 



Nevertheless, he proposes for Paheorhiza a distinct 

 family, Palaeorhizidae, at the same time suggesting that 

 it should be Meroglossidae if Meroglossa is allied. 



In the course of going over the splendid collection of 

 .Australian bees formed by Mr. Rowland E. Turner, now 

 the property of the British Museum, I have been able to 

 examine the structure of a number of species of Palaeo- 

 rhiza. In the first place, I find that Palaeorhiza and 

 Meroglossa are substantially the same genus ; but the truly 

 astonishing thing is that the females have broad, obtuse 

 tongues like Prosopis, while the males have sharp, dagger- 

 like tongues ! I first discovered this in P. penetrata, 

 subsp. percrassa (properly Meroglossa penetrata percrassa), 

 a black insect with the face of the male canaliculate, much 

 in the manner of the original type of Meroglossa. Mv 

 natural thought was that there must be two species, in 

 spite of every appearance to the contrary. I next took 



