14 DIMORPHIC BRANCHES IN TROPICAL CROP PLANTS. 



or left-handed with reference to the position in which their fruiting 

 branches are borne along the main stalk. 



On the fruiting branches this regularity in the position of the 

 flowers is not so obvious, for the joints are twisted to bring all the 

 leaves to the sides and all the flower buds on top. The flower buds 

 appear between the bases of the stipules, sometimes nearer the right- 

 hand stipule, sometimes nearer the left. The stipule that is close to 

 the base of the flower stalk is usually larger than the other. 



Antidromy, as the condition of right and left handedness of plants 

 has been called, consists in the fact that the stems of the different indi- 

 vidual plants reverse the direction of the spirals in which the leaves 

 and branches are arranged. On some cotton plants the extra-axillary 

 branches occur on the right side of the axillary branches; in other 

 individuals on the left side. If a stalk on which the extra-axillary 

 buds appear to the right of the axillary buds be considered right- 

 handed, the turn of the spiral will pass to the right in going by the 

 shortest route from any given branch to the one above. Thus it 

 appears that the extra-axillary bud is always above the axillary, in 

 the sense that it is farther up the spiral. 



In all the different species and varieties of cotton thus far exam- 

 ined right-handed and left-handed stalks seem to be about equally 

 numerous. As the Guatemalan types in which the branch dimor- 

 phism was first studied had never undergone close selection, the ques- 

 tion was raised whether among the carefully bred American varieties 

 there might not be specializations toward one direction of the spiral. 

 No indication of this was found in a large series of varieties studied 

 by Mr. F. J. Tyler at Waco, Tex. Seeds from the same boll were 

 also found to give about equal proportions of right-handed and left- 

 handed seedlings. The possibility remained that the direction of the 

 spiral may be determined in advance by the positions in which the 

 seeds develop on the placenta. To test this theory seeds from two 

 rows of the placenta were planted separately, but gave right and 

 left handed plants without reference to the position on the placenta. 

 The manner in which this diversity arises remains unexplained. 



The axillary buds have been found in all the types and varieties 

 of cotton thus far observed, but they are often very small and 

 dormant. They may all remain undeveloped unless the plant is cut 

 back or severely checked by unfavorable conditions. In many kinds 

 of cotton both types of branches are commonly to be found on the 

 same plant. 



Tlie difference between the two kinds of branches was first appre- 

 ciated in Guatemala in connection with the indigenous Kekchi cotton. 

 The lower joints of the main stem of the Kekchi cotton usually pro- 

 duce two branches, one a fertile branch with flowers and fruit, the 



198 



