DR. I. B. BALFOTJK ON THE GENUS PANDANUS. 33 



Observations on the G-enus Pandanus (Screw-Pines) ; with an 



Enumeration of all Species described or named in Books, 

 Herbaria, and Nurserymen's Catalogues; together with 



their Synonyms and Native Countries as far aMrfrese have 

 been ascertained. By Dr. Isaac Bayley Balfour, F.L.S., 

 F.R.S.E., &c. 



[Read December 6, 1877.] 



Few families of plants present more difficulty in their elucidation 

 than the Pandanacese. This arises from a threefold cause : — in the 

 first place, from the variability of the species, dependent not only 

 on relative position with reference to climatic influences, but also 

 on the age of the individuals. Secondly, we find it is by no means 

 easy to obtain characters sufficiently diagnostic, as the leaves 

 afford marks of little or no value ; and hitherto it has been from 

 the fruit that distinctions have been drawn. The flowers of the 

 male tree yield in some instances good characters ; but we have 

 as yet received flowers of only a few species, and these in some 

 cases without information sufficient for identification with the 

 female. Lastly, in dealing with primitive types of such varia- 

 bility, it is necessary to obtain a long series of specimens ere 

 we can determine with certainty the limits of species ; and these, 

 I regret to say, are still in great measure wanting. The diffi- 

 culty regarding specimens is still further increased by the fact 

 that the fruits lose much of their character in drying. It is 

 therefore almost impossible to investigate the group without some 

 experience of the plants in their native haunts, where, by a con- 

 sideration of the combination of characters derived from habit, 

 foliage, flower, and fruit, a correct estimate may be formed. 



Having had opportunities of studying the group as repre- 

 sented in the Mascarene Islands, and having already described* 

 some species from these islands and from the Seychelles Islands, 

 I have been tempted to enter on a large undertaking, and to 

 contemplate a monograph of the whole Order. With this 

 view I now venture to lay before the Society a synopsis of the 

 species of the genus Pandanus, so far as I have been able to 

 determine them. I am quite conscious that, as it now stands 

 the list is a very imperfect one, and there are numerous errors. 

 It can hardly be otherwise, founded, as it is, so greatly on mere 

 descriptions usually very unsatisfactory. But I am advised to 

 bring it forward in its present form, with the hope that any persons 



* Baker's ' Flora of Mauritius and Seychelles,' p. 395. 



LINN. JOTJEN. — BOTANY, TOL. XTII. D 



