288 THE JOURXAL OF BOTAXT 



the plants to Kew in 1901 for identification and description. This, 

 task was given to me and I furnislied a report on the same in 1902 

 for jniblication ; but I believe it has never appeared. 



Dioscorea nesiotis Hemsl. {Dioscoreacecs) ; species ex affinifcite 

 D. Tiisonl Baker (Africie australis incola), a qua differt omnino 

 glal)ra, iloribus minoribus. 



Herba tuberosa undique glabra, eaulibus scandentibus gracillimis 

 monocarpicis. Folia alterna, cum petiolo gracili usc^ue ad 10-12 cm. 

 lon.'Lra, 3-'3-f()liolata ; foliola tenuissima, subsessilia, ovato-oblonga, 

 2-8 cm. longa, acute acuminata, venis inconspicuis. Flores uni- 

 sexuales, ut videtur, dioici, minimi, simpliciter racemosi, racemis axil- 

 laribus solitariis vel interdum binis breviter pedunculatis, masculinis 

 quam femineis brevioribus axi per anthesin fere ca])illari : llores mas- 

 culini distinete pedicellati pedicellis bracteis a^quilongis acutis sub- 

 tendis. Perianthii segmenta fere libera, circiter 1-25 mm. longa, 

 ovali-oblonga, apiculata. Stamine 6, filamentis brevissimis. Flores 

 feminei perfecti non visi. Perianthii segmenta oblonga 1-5-2 mm. 

 longa. Kncenii fructigeri 15-20 cm. longi, densi. Capsula 3-alata, 

 alis fere semiorbicularibus circiter 1 cm. latis. Semina orbicularia, 

 compressa, circumcirca ala membranacea cincta. 



Aldabra: Thomasset, 21.S, 211. 



The same species was collected in Assumption by Dupont, 118. 



BIBLIOGRAPHICAL NOTES. 

 LXIX. The PAGE-HEADiNa of Periodicals. 



Ix three of the earlier notes in this series (Journ. Bot. 1891, 180, 

 271; 1896, 168) 1 dealt with certain points in connection with the 

 dating and indexing of periodicals which seemed to demand attention. 

 The suggestions made in the two former of these were, as I w^as able to 

 sav when writing the last of them, in so many cases acted upon that 

 ] am venturing now to add a note on another matter relating to 

 l)eriodicals which may perhaps prove equally useful. 



At the outset, as in the previous cases, it would seem that no 

 suggesti(Mis could be necessary : it might have been expected that 

 custom woukl have long since decided upon the most convenient and 

 most useful way of indicating the contents of a periodical, and that 

 there was little room for difference of opinion and of practice. This 

 would seem to be especially true in the case of page-headings ; and 

 the fact that even in so simple a matter so great a variety exists can 

 only be explained by the natural tendency of the human mind to 

 variation, for it can hardly be supposed that the divei*sity is inten- 

 tional. 



As has been more than once pointed out in this Journal in the 

 course of l)ook-reviews, the bUmk space at the top of each page, 

 necessitated by the page-number, afl'ords an opportunity for con- 

 veying information as to wdiat stands below it. An inspection of 

 botanical ])eri()dicals printed in English, with which alone this note is 

 concerned, shows that in many instances this o])])ortunity is wholly or 



