390 



KNOWLEDGE & SCIENTIFIC NEWS. 



[March, igo6. 



loosely attached to the atoms and not apparently forming an 

 integral part of them. They seem to float about very much 

 as the particles of gas dissolved in a liquid. The electrons 

 which leave the substance under the action of Kontgen rays 

 may simply be these unattached ones, and if so the existence 

 of the streams is no proof of atomic disintegration. This 

 (piestion has recently been tested by Professor Bumstead 

 (working in the Cavendish Laboratory), who argues that if 

 disintegration is brought about there must be a corresponding 

 release of atomic energy, which will appear — in the main — 

 as heat, and which will raise the temperature of the substance. 

 For equal absorption of Rontgen rays the amount of heat 

 produced from two substances might be expected to depend 

 upon the material and be greater in amount than that of the 

 incident rays, which merely play the part of a trigger. But if 

 no disintegration occurs the heat produced should be equal 

 to the energy- of the incident radiation (minus the small 

 amount carried away by the secondary radiation), and should 

 therefore be practically the tame in both cases. The experi- 

 mental method employed is based upon these considerations. 

 It is in brief a sort of reversed radiometer. A radiometer 

 — such as was first devised by Crookes — acts owing to the in- 

 cident light warming the vane which is repelled by the forces 

 in tlu-iiiy between the vane and the fixed wall of the tube. It 

 comes to the same thing if the wall itself is warmed while the 

 vane is made transparent so as to keep cool. The latter 

 arrangement was most suitable in the present case. The walls 

 can be made of the metal under investigation, and the move- 

 ment of the suspended vanes is due to the warming of the 

 walls by Rontgen rays, and is a measure of the heat 

 produced. 



Only lead and zinc have been experimented upon so far. 

 Their thicknesses were chosen so that they absorbed the 

 same amount of radiation, and under these circumstances the 

 experimental result is that about twice as much heat is gene- 

 rated in the lead as in the zinc. 



Professor Bumstead examines the numerous sources of 

 error to w^hich the mode of experimenting is subject ; and also 

 other modes of explaining this difference in heat production ; 

 but concludes that the evidence is in favour of the view that 

 true atomic disintegration is brought about by the absorption 

 of Rontgen rays. The experiment is, however, beset with 

 pitfalls, and physicists will probably await the results of 

 further experiments, which are promised, and in which it is 

 proposed to employ other means of measuring temperature, 

 such as a thermopile. Anyone acquainted with the vagaries of 

 a radiometer will regard with caution the conclusions arrived 

 at from these preliminary experiments, even though he may 

 be willing to grant their plausibility. 



(" Philosoi'HIcai. Maga^ink." Feb. njo6.) 



Conductivity of the Vapour from a. 

 Mercury Arc. 



The Hon. R. J. Strutt showed some years ago that mercury 

 \apour under ordinary conditions is a very perfect non-con- 

 ductor of electricity. C. D. Child, in a recent paper read 

 before the American Physical Society, shows that the vapour 

 coming from a mercury arc is highly conducting. A lumi- 

 nous space gradually spreads out from the arc, and the front 

 of this region has the greatest conductivity and the greatest 

 luminosity. He considers that the conductivity is not due 

 directly to ions coming from the arc, nor to rays sent out by 

 it, nor to leakage over the surface of the glass, and probably 

 not to the high temperature of the gas. He suggests as prob- 

 able an explanation given by Merritt, viz., that when ions 

 recombine they are at first in a condition of unstable equilib- 

 rium much like the atoms of radioactive matter, and that 

 many of these combinations break up again into positive and 

 negative ions. 



The Efficiency of a Welsbach Ma.ntle. 



A fresh study has been made by Rubens of the light from a 

 Welsbach mantle with the object of elucidating the reason of 

 the extraordinary efticiency of this source of light. The fact 

 that the small addition of cerium oxide (-8 per cent.) to the 

 thorium oxide which forms the main body of the mantle 

 ((ji)-.; per cent.) enormously increases its light-giving power has 

 induced many— especially among chemists — to regard the light 

 as being essentially of phosphorescent origin — i.e., as depend- 

 ing in the main upon chemical changes (or catalytic action) 

 rather than upon the high temperature to which the mantle is 



raised. It is well known that at a given temperature, the 

 radiation which depends upon temperature can never exceed 

 that from a " perfectly black body " at the same tem- 

 perature. Since a mantle without ceria and one with it give 

 out very different light wlien placed in the same Bun sen flame, 

 the assumption that the additional light is not due to the tem- 

 perature has at least some plausibility. But nothing except a 

 thorough study of the radiation can suffice to settle this in- 

 teresting question. For there is a tacit assumption amongst 

 the holders of the chemical theory that the two mantles placed 

 successively in the same flame will have the same temperature. 

 It is clear, however, that if one radiates much more than 

 the other it will necessarily adjust itself to a /oii'ic tem- 

 perature ; for it is the gradient of temperature from the 

 flame to it which controls the flow of energy to the mantle 

 and which maintains this mantle at a constant temperature 

 in spite of its radiation. The one that radiates most has to 

 receive most, and a greater gradient is necessarily required to 

 keep up the supply. Those who claim that no chemical 

 change is responsible for the light assert that the more lumi- 

 nous mantle owes its luminosity to the fact that at a given 

 temperature it is a bad radiator for that tcmperciturc ; but that 

 owing to this parsimony it is able to keep at a higher tempera- 

 ture than its more generous companion ; and everyone is 

 agreed that a higher temperature tends to produce greater 

 efficiency. 



The upshot of Rubens' measurements is that a mantle 

 stained with iron oxide (and therefore a good radiator at a 

 given temperature) kept at atemperature of about 1050" C. in 

 a flame ; whereas the unstained mantle rose to about 1500 C, 

 while the estimated value of the temperature of the bunsen 

 flame itself is about iSoo° C. After making allowance for the 

 non-continuous nature of the surface of the mantle, it is shown 

 that /or no wave length is the radiation as great as that from a 

 perfectly black body at the temperature of 1500^' C. ; and that 

 the total energy radiated is only one thirty-third part as great 

 as from such a body. .Although then it cannot be denied that 

 phosphorescence may play some small part, yet it is altogether 

 unnecessary to call in its aid to explain the amount of radia- 

 tion. 



Experiments on a mantle of pure thorium oxide show that 

 the radiation from it contains almost no light rays, but in other 

 respects is very nearly the same as for an ordinary mantle. 

 On the other hand, a mantle of pure cerium oxide gives out 

 radiation proportionately very rich in luminous radiation, and 

 also of radiation of very long wave length. Vox the important 

 region of wave lengths immediately in the infra-red {i.e., X = i 

 to 8-millionths of a metre) it is a bad radiator. It would not 

 do, however, to use such a mantle, for its temperature (like the 

 iron-oxide one) keeps very low. But the addition of a small 

 amount of ceria to thoria much increases the luminous radia- 

 tion of the latter without sensibly alterating the radiation in 

 the immediate infra-red. Thus the behaviour of an ordinary 

 mantle can be adequately represented by adding together the 

 separate radiations of the constituents of the mantle. The 

 opposite supposition, that the light is due to phosphorescence, 

 requires that the greater part of the effect should arise only 

 when the constituents are present together. This result 

 seems to be in conflict with the new experimental data. 



ZOOLOGICAL. 



i!y K. LVDEKKEK. 



The Intestinal Appendages of Birds and 

 Ma^mma-ls 



-An important contribution to our knowledge of the intestinal 

 anatomy of mammals has just been published in the Transac- 

 tions of the Zoological Society, Dr. Chalmers Mitchell, the 

 Society's secretary, being the author. Perhaps the most 

 generally interesting feature in this memoir is the identifica- 

 tion of the paired •' c;eca," or blind appendages of the intes- 

 tine, of birds with the, usually, single ca-cum of mammals. 

 These ca;ca occur at the junction of the small with the large 

 intestine ; and while in ordinary perching birds they are 

 reduced to small nipple-like buds of no functional importance, 

 in many other birds — owls for instance — they form quite long 

 receptacles. Among mammals, the horse and the dog may 



