566 



NATURE 



[December 29, 192 1 



both pla\' similar roUs in the corcnionics, and ap- 

 parently both equally need excitation for coition to 

 take place. Here mutual, instead of one-sided, 

 excitation occurs. 



It will be evident that wherever mating displays and 

 ceremonies, and the colours and structures associated 

 with them, have this purely stimulative function they 

 cannot be supposed to stand in any relation to sexual 

 selection, but resemble copulatory organs in being 

 solely subservient to eflficiency in securing union of the 

 gametes. Copulatory organs will arise only when 

 a certain level of general physical complexity is 

 reached ; stimulative displays, colours, and structures 

 only at a certain level of sensory and nervous (mental) 

 complexit}-. 



The problem of their genesis is therefore no more 

 difficult than that of copulatory organs, or, indeed, of 

 any other adaptive structures — which is not to sav 

 that it is easy, but that at least it does not demand 

 special evolutionary agencies. 



When polygamy obtains, true sexual selection mav, 

 of course, occur, and from observations such as those 

 of Selous on the ruff and blackcock (Selous, 

 Zoologist, 1906 and 1909-10), does definitely seem to 

 be operative. 



To denote characters which are secondarilv con- 

 cerned wnth bringing about mating or copulation we 

 may use the term "epigamic," first coined by Poulton. 

 It is necessary to have such a term, since "secondary 

 sexual " cannot be applied to epigamic characters 

 occurring in both sexes, and can be applied to non- 

 epigamic characters such as mammae. Characters 

 effective in the act of copulation itself (copulatorv 

 organs) come under the head of "accessory sexual 

 characters," Epigamic characters are then either 

 "accessory," when they stimulate to coition, or 

 "sexually selected," when they stimulate to the selec- 

 tion of mates. In certain cases thev mav be both. 

 There is, however, good reason for believing that the 

 great bulk of sexual displays, with their associated 

 cofours and structures, are accessory epigamic char- 

 acters, and that the problem of their evolutionary 

 origin is therefore simplified. 



Julian S. Huxley. 



New College, Oxford, December 7. 



Terrestrial Magnetic Disturbances and Sun-spots. 



Referring to Father Cortie's letter on this subject 

 (Nature, October 27, p. 272), the sequence of mag- 

 netic disturbances following at 27-day intervals the 

 storms of May 12-21, 192 1, was recorded also at 

 Kodaikanal, but those storms recorded in England as 

 "very great " on September 2 and 28-29 ^vere classed 

 here as "moderate " only. There were also recorded 

 here "great " storms on March 21-22 and April 18-19, 

 which appear to belong to the same sequence, the 

 whole interval from March 22 to September 29 giving 

 a mean period of 2729 days. Assuming this to be 

 the synodical period of the sun, the equivalent sidereal 

 period is 2542 days, closely agreeing with Carring- 

 ton's mean value of the rotation for spots. 



Our spectroheliogram records show that the spot 

 disturbance originated on the invisible hemisphere of 

 the sun shortly before May 8, when it first appeared 

 on the east limb as a condensed mass of bright 

 calcium flocculi. As is usually the case, the area of 

 disturbance became extended ; in July the flocculi 

 were more or less scattered, and the last visible 

 remnants of these are seen in the photographs of 

 August 1-8. 



The magnetic disturbances seem therefore to have 

 both preceded and followed the spot disturbance as 



NO. 2722, VOL. 108] 



recorded in calcium light, the flocculi lasting through 

 three synodic rotations and the magnetic effects 

 through seven, 



E.xcept during May, there were no striking 

 manifestations of prominence activity recorded here 

 near this group of spots. On May 9 we photographed 

 on the east limb a group of circular interlacing arches 

 such as are occasionally seen over newly formed 

 spots, and during the passage of the group across the 

 visible hemisphere there were very brilliant metallic 

 reversals and other manifestations of eruptive activity, 

 besides innumerable absorption markings in Ha light 

 intertwined throughout the group. All this activity 

 appears to have subsided during the following month. 



.Another more striking instance of the persistence of 

 magnetic efi'ects long after every trace of solar dis- 

 turbance has disappeared occurred in the year 1920. 

 The "very great " storm of March 22, 1920, was asso- 

 ciated with a very large spot group passing the central 

 meridian on that date. This group and its associated 

 flocculi were observed during five synodic rotations 

 from January i to April 18, and each meridian pas- 

 sage was accompanied by a magnetic storm of 

 "great," "very great," or "moderate" intensity. 

 The spot disturbance and its surrounding flocculi com- 

 pletely subsided during May, yet the magnetic disturb- 

 ances continued to recur at 27-day intervals for seven 

 more periods ; the last disturbance to be identified in 

 this sequence was recorded on November 21 as a 

 "moderate" storm. The interval January i to 

 November 21, 1920, is 325 days, or twelve periods of 

 27-08 days. Allowing for the earth's orbital move- 

 ment during this interval the equivalent solar period 

 is 2522 days, or Carrington's rotation period for spots 

 in latitude 10°. The slight difference of period com- 

 pared with that obtained from the 192 1 series does 

 not make the evidence for these sequences less con- 

 vincing ; in his discussion of a very large number of 

 such sequences Maunder has shown that the sidereal 

 period derived from them may vary from 25 days to 

 265 days (Monthly Notices, R.A.S., vol. 65, p. 553). 



J. EVERSHED. 



Kodaikanal, November 28. 



Microscope Illumination and Fatigue. 



The further letter from Mr. H. J. Denham in 

 Nature of December 15, p. 496, does not in any way 

 alter my opinion that the method of varying the 

 intensity of illumination in the microscope described 

 by him is not the best or the most convenient at 

 present available. The use of a monochromatic light- 

 filter does not affect the question, as such an acces- 

 sory is used or not as may appear desirable in anyj 

 given circumstances. At this institute several sources 

 of light are installed, the one that is regarded as thel 

 most useful in high-power work being the mercuryj 

 vapour lamp. It is obvious, therefore, that if light- 

 absorbing screens of known opacity are available,! 

 nothing further is needed whether the light is mono- 

 chromatic or otherwise. .\s I have already stated, 

 such screens as we use here alter the intensity, anc 

 not the character, of any visible light which the\ 

 transmit. If Mr. Denham regards the change ofj 

 quality of his light as an advantage, I do not thinl 

 that many microscopists will agree with him or adopt 

 his methods while more efficient ones are at hand. 



As to mv experience of the value of variable illu-j 

 mination, it is, in mv opinion, an essential feature ofj 

 any good microscopic outfit. In use it is second onlyf 

 in 'importance to the proper adjustment of the light 

 and sub-stage condenser. .Apart from its advantage 

 to the observer in reducing fatigue, it is often possible 

 to use a larger cone of illumination by reducing the 



