174 



NA TURE 



[June 2f, 1900 



comparative feebleness of the shower in this month. An observer, 

 however, who extends his watch over a long period, if not over 

 the whole of the night, will find little difficulty in mappmg a 

 sufficient number of Perseids to indicate a good radiant. 

 Bishopston, Bristol, June lo. W. F. Denning. 



Variations in Plants of the Herb Paris. 

 The enclosed table, showing the variations in 200 plants of 

 Herb Paris, picked this month in the woods near Wells, may 

 be of interest to some of your readers, especially if looked at in 

 connection with the memorandum written by Sir Edward Fry, 

 which he is kind enough to allow me to send with it. 



L. Eleanor Jex-Blake. 



[Miss Jex-Blake's table seems to me to show many points of 

 interest. 



The Herb Paris has long been known to be very variable in 

 the number of its parts ; this table quantifies (I use the word, 

 though it used to make a friend of mine very angry) the vari- 

 ability of the plant. It shows that, taking the 96 plants as 

 exhibiting the normal form, more than one-half, i.e. 104 out of 

 200, vary from the standard ; that the most variable element is 

 the circle of stem leaves ; and that looking at the flowers alone, 

 142 plants out of 200 are normal, 58 only abnormal ; that the 

 58 thus varying plants fall into no less than 28 groups ; that 

 not only do the plants vary as wholes, but that parts usually 

 the same in number, or multiples of the same number, do not 

 maintain this relation, e.g. that in 13 plants you get 5 sepals, 

 4 petals and 9 stamens, and so on. 



The plant being thus given over to variability and belonging 

 to the great group of monocotyledons, in which 3 and miilti- 



NO. 1599, VOL. 62] 



pies of 3 are the dominant number for the parts of the flower, 

 a systematist might expect that the variations of the Herb Paris 

 would oscillate round 3, or a multiple of 3, as the standard 

 form ; but, in fact, they oscillate round 4 as the dominant 

 number, the 96 normal plants having that number, or a multiple 

 of that number, everywhere, and 44 plants having that number 

 and multiple everywhere except in the leaves. Nature, there- 

 fore, disappoints our reasonable expectation. 



It has, I believe, been suggested that the flower of Herb 

 Paris is ideally of 6 and 12 parts, and that it has been reduced 

 to 4 and 8 parts by atrophy and suppression of 2 and 4 parts 

 respectively. If this were a true theory, you would expect to 

 find here and there a reversion to the ancestral form ; but the 

 table shows that the number 6 occurs in the floral parts once, 

 and once only, viz. in the sepals, and the number 12 never 

 occurs in the stamens or elsewhere, so that the suggestion of 

 suppressed parts becomes highly improbable. 



The Herb Paris wanders from the ordinary type of mono- 

 cotyledons, not only in the number of the floral parts, but in 

 having ramifying veins of the leaves in the place of parallel 

 veins ; there are other monocotyledons which have this variation 

 in the leaf from the standard. Do they, too, show any 

 tendency to vary in the number of the floral parts ? or to put it 

 in other words, is there any correlation of the two variations ? 

 I have not looked into the subject, but it might prove worth 

 consideration. — E. F.] 



May 25. 



Quaternion Methods applied to Dynamics. 



I SHALL be obliged if any of your readers can give me the 

 titles of any works on statics, or dynamics, or any physical 

 science which are based on Quarternion methods and use nothing 

 but Quaternion symbols. 



The end chapters of P. G. Tait's " Quaternions " give examples \ 

 Kelland and Tait work out the theory of strains using Quater- 

 nion methods, but neither of these suffice for the purpose I have 

 in view, namely, to put into the hands of a student a text-book 

 on dynamics, &c., written in Quaternion language. 



Jubbelpore, June i. W. G. Barnett. 



H 



PLANT HYBRIDS. 

 ORTICULTURISTS have recognised that as 

 time goes on they must look more and mora to 

 hybridisation for "new plants.'' Biologists are already 

 pointing out that, if anything can, breeding experiments 

 will add to our knowledge of "the species." For both of 

 these reasons the current volume ^ of the Royal Horti- 

 cultural Society's Journal is of very particular interest, 

 seeing that it is in fact the detailed report of the Confer- 

 ence on Hybridisation and Cross-breeding held last 

 summer. The present writer has already summarised in 

 these pages ^ the chief facts of importance brought out in 

 the two days' proceedings ; but several of the papers have 

 been elaborated and illustrated, while many further con- 

 tributions have been sent in and are now published. The 

 latter in particular call for further comment. 



Speaking generally of the report, it may be said that it 

 is of very great value as a record of parentage, as a store- 

 house of many facts, and as putting forth several inter- 

 esting theories. Furthermore, among the contributors 

 are amateur and professional horticulturists, scientific 

 workers pure and simple, as well as men who combine 

 the interests of both, and this is a decided step in the 

 right direction. It is not to be expected that the collec- 

 tion of papers forms a complete treatise to guide the 

 practical or theoretical student ; useful points are only to 

 be found among cases at present not to be reconciled 

 together and along with striking differences of opinion. 



The very discrepancies are, however, to be welcomed, 

 for from them can be learned the work to which attention 

 should be most ungrudgingly given in the future ; and by 

 the publication of the " Hybrid Conference Report" the 

 Royal Horticultural Society will earn the gratitude of a 

 larger circle than ever. In the present account it will be: 



^ Jonrn. R.H.S. \o\. xxiv. (April 1900), pp. 1-34S ; 123 Figs. 

 2 Nature, vol. Ix. (No. 1552, July 27, 1899), pp. 305-307. 



