204 



NATURE 



[April 14, 192 1 



When the excitation is produced by Di, the ratio of 

 intensities of the two resonating Hnes Dj and D,, 

 which is very small when the temperature and density 

 are low, rapidly increases to its normal value 2 with 

 the number of atomic collisions per second. This 

 number is thus given statistical significance. Until 

 now there has been no strong evidence (Wood, 1914) 

 as to its invariability at higher temperatures and 

 pressures. 



Special investigations on this subject made in 1915 

 and 1917, and published in Russia, appear to be un- 

 known abroad. The dispersion of the vapour of 

 alkali metals was studied in 1915. For all the first 

 doublets of Na, K, Rb, and Cs the same value 2 was 

 obtained, and it remained constant in spite of a 

 hundredfold density variation ; for the second doublets 

 the numbers are simple, but different: 2(?), 2, 2-5, 4. 

 The numbers 3 and 7 (?) were measured for the 

 third doublets of Rb and Cs. Mr. Touroverow (1917) 

 found the same number 2 for the first sodium doublet 

 at the temperature of the arc. There is, therefore, 

 no doubt now as to the constancy of all the above 

 numbers. The experiments on resonance thus show 

 that the statistical value in question first grows 

 rapidly with the temperature and approaches a limit- 

 ing value, essentially constant. This behaviour has 

 a certain analogy to that of specific heat as caused 

 by departure from equipartition. 



D. ROGESTVENSKY. 



Petrograd University Physical Institute, 

 March. 



The Resonance Theory of Hearing. 



The discussions which appeared in Nature in 19 18 

 (vol. cii., pp. 124, 164, 184) on the theory of hearing 

 showed that the opinion has been gaining ground 

 lately that the resonance theory can no longer be 

 regarded as unassailable. The following observation, 

 which is readily explicable if there are resonators in 

 the internal ear, would appear to be inexplicable if 

 there are not : 



If the phase of a continuous musical note be sud- 

 denly altered by suitable means by r, then the ob- 

 server hears the sound rapidly die away, to return a 

 moment later with its former intensity. The experi- 

 ment was performed as follows : 



A De la Tour siren was so modified that the wind- 

 chest could be given suddenly a small rotation about 

 the same axis as that of the siren disc. The rotation 

 was limited by fixed stops, so that the angle turned 

 through was equal to one-half the angle between two 

 of the air-holes. In the writer's instrument there 

 were eighteen holes arranged on the circumference, 

 i.e. 20° between two of the holes, and the wind-chest 

 was therefore arranged to rotate through 10°. If, 

 then, this rotation is suddenly effected with the siren 

 In action, a change in the phase of the note of n- 

 will be introduced ; since, if the rotation of the wind- 

 chest be in the same direction as that of the disc, 

 the time-interval between the puffs of wind through 

 the disfi will be \\ times as great as the normal, 

 because the disc has to rotate through 20°+ 10° ; if, 

 on the other hand, the rotation of the wind-chest be 

 in the opposite direction to that of the disc, the time- 

 interval will be one-half the normal, since flie disc 

 has to rotate through 20°— 10°. Each time, then, 

 that this change of phase of it is brought about bv 

 rotation of the wind-chest of the siren the observer 

 hears a beat in the musical note. The sound intensity 

 first falls to a low value, then rapidly rises somewhat 

 above the original level (possibly due to successive 



NO. 2685, VOL. 107] 



contrast), and then returns and stays at the normal 

 intensity. To show that the beat is not of mechanical 

 production the following tests may be applied : 



(a) No beat is produced if the wind-chest is rotated 

 slowly. 



(b) No beat is produced if, with the disc in rotation, 

 the air-supply be quickly turned off and the wind- 

 chest then rotated suddenly in either direction. 



(c) The beat can be heard as clearly at a consider- 

 able distance from the instrument as it can quite near 

 to it. 



(d) If the rotation of the wind-chest is less than 

 that required to change the phase by tt, the beat or 

 temporary waning of the note is correspondingly 

 smaller in intensity. 



This temporary waning of the note is .readily ex- 

 plained by the resonance theory, because the change 

 in phase will put the later vibrations exactly out of 

 step with those that preceded, and therefore the 

 resonators of the internal ear which are set in vibra- 

 tion bv the note will on change of phase first be 

 brought to rest and then be set going again. The 

 temporary waning of the note is therefore readily 

 explained on the resonance theory. Can any of your 

 readers advance an explanation on any of the dis- 

 placement {e.g. Wrightson's) hypotheses of hearing? 



H. Hartridge. 



King's College, Cambridge, March 21. 



Sexual Organs of Phytophthora. 



Attention was directed in Nature of April 30, 

 1914 (vol. xciii,, p. 226), to the discovery of a rather 

 remarkable mode of development of sexual organs 

 which occurs in certain species of Phytophthora, and 

 was first found in P. erythroseptica and then in P. 

 infestans, the "potato-blight" fungus. Several other 

 species of the genus are now known to produce sexual 

 organs in this novel fashion, in which the oogonial 

 incept penetrates the antheridium at an early stage, 

 traverses it, emerges, and then swells to form the 

 oogonium proper within which the oospore ultimately 

 develops. It was suggested then that those previously 

 well-known species (such as P. caciorum, etc.) in 

 which the antheridium and the oogonium lie side by 

 side, and penetration of the latter by the former 

 occurs laterally, should be excluded from the 

 genus Phytophthora and be placed in a new one, 

 Nozemia. A species (from decaying apples) has now 

 been isolated by Mr. H. A. Lafferty, working here, 

 in which the sexual organs are developed mainly 

 according to the Nozemia type, but occasionally and 

 simultaneously in the same individual according to the 

 Phytophthora type, with amphigynal antheridia. This 

 species, therefore, forms a connecting link between 

 the two groups ; and it would seem no longer neces- 

 sary or desirable to retain the generic name Nozemia. 



The object of this letter is to suggest to the various 

 mycologists who are now working with Phytophthoras 

 that they should keep a very careful look-out in cul- 

 tures of species of the Cactorum or omnivora 

 (Nozemia) type for the occasional occurrence of sexual 

 organs with amphigynal antheridia ; for it seems quite 

 possible that these may be present in such species and 

 have merely been overlooked by previous observers 

 owing to their relatively infrequent occurrence. 



I should be very grateful for sub-cultures of any 

 species of Phytophthora that mycologists who have 

 them could spare for further study of this point, and 

 happy to send any I possess in exchange if desired. 



Geo. H. Pethybridge. 



Royal College of Science, Dublin, April 7. 



