1 38 BULLETI N J A RD. BOT. BUITENZQRG, SÉRIE lli, VOL. V. LiVR. 2. 



Summary. 



The oil palm ordinarily produces nutritive and adventitious roots. 



Characteristic air chambers or lacunae are found in ail the roots. 



Thèse chambers origiiiate in the radially arranged parenchymous tissue 

 of the periblem. 



Thèse chambers arise as fissures or slits in the tissue apparently through 

 actual tearing of cells. 



Two kinds of secondary roots are developed ordinary nutritive and 

 pneumathodes. 



There are no root hairs formed on any of the roots. 



Pneumathodes are secondary roots with a single bundle running through 

 the center and with specially developed aerenchymous tissue. 



The fruit. 



The gross anatomical structure of the oil palm fruit has served as the 

 basis of classification of the many so-called varieties. Other characters of 

 the tree hâve been employed in a lesser degree for that purpose. It is 

 readily understood that a plant so extensively used by the indigenous popu- 

 lations, of Africa primarily, v^ould receive native names based on obvious 

 différences. The fruit of the plant being the part used by the natives, 

 variations in the fruit then, hâve naturally been used for varietal classi- 

 fication. In the fruit the following distinguishing characters hâve been used: 

 Color of fruit before and after ripening including présence or absence of 

 anthocyan and chlorophyll. 



Form of fruit 



Thickness of shell — thick or thin shell 



Absence of shell 



Présence of accessory stérile carpels. 



Varieties of fruits. 



As a resuit of thèse observations a very large synonomy has sprung 

 up — the natives of a given locality naming the forms found growing there. 

 The work of the investigators who hâve attempted to systematize the forms 

 has unfortunately led to no satisfactory classification, at least not from the 

 point of view of the student of genetics. That thèse différences exist cannot 

 be denied. However no work has been done to ascertain to what extemt 

 such différences may be due to variations within a given line, how stable 

 those différences are, what rôle environment plays, and what can be 

 ascribed to teratological and to pathological phenomena. My investigations 

 in that direction C^) which hâve served as the basis of a small contribution 

 and for the basis of a more extensive one to appear later hâve led 

 me to question the diagnostic value of many such characters. 



