54 WEEVIL-RESISTING ADAPTATIONS OF COTTON. 



half Specked with minute liroAvn glandular dots." The oil glands of 

 the bolls of this Kidney cotton are apparently quite as strongly devel- 

 oped as in the Egyptian varieties, or even more so. Thev are distrib 

 iited very irregularly over the surface, and are not lacking above the 

 dissepiments, along the middle of the carpels. The position and 

 structure of the glands seem also to be the same as in the Egyptian 

 cottons. They are close to the surface and show as distinct black 

 spots, there being no green tissues over them as in the Upland and 

 Jterbaceum types. 



I am indebted to Mr. Guy N. Collins for the suggestion that the 

 present inefficiency of the oil glands as a means of protecting the 

 cotton from the boll weevil furnishes no argument against the 

 adaptation of the glands nor their development through the selective 

 agencies of the boll w^eevil itself. This fact is sufficiently obvious 

 when once stated, but it is not commonly taken into account in con- 

 sidering questions of this kind. We may be sure that the gradual 

 development of a protective character like the oil gland would carry 

 with it a corresponding increase in the power of the weevil to avoid 

 or to endure the injury. The ultimate value of the device would de- 

 pend on whether the glands were able to keep ahead of the Aveevils 

 in quantity and distastefulness. The readiness with which the boll 

 weevils attack the Egyptian cotton renders it obvious that oil is now 

 no adequate protection, but the preference of the weevils for the un- 

 protected strips of the bolls of the Kekchi cotton indicates that the 

 weevils still dislike the oil, though they may have foiled the attempt 

 of the plant to protect itself in this way. 



There are tAvo attendant facts which under certain circumstances 

 might readily obscure the immunity of the young bolls. Many such 

 small bolls fall off, a particularly large number it seemed from our 

 row of Parker cotton, but an examination of these failed to show 

 anything in the way of weevil injuries, except such as had been in- 

 flicted wdiile the bud or flower was still in place, the style and a 

 small apical cavity having been eaten away in numerous instances. 

 Many small bolls w^ere to all appearances quite uninjured. They 

 may have been rejected by the plant as supernumerary, the plant 

 being unable to furnish the food material needed to bring them to 

 maturity, or they may have failed of fertilization as a result of 

 w'eevil injuries to the bud or from other causes, such as the absence 

 of bees, w-hich were extremely scarce in the Guatemalan cotton fields. 

 The frequency with which the boll weevils were found inside the 



a The flowers of the Kekchi cotton are pure creamy \Ahite when young and as 

 long as they remain open. When old and rolled together they become a pinkish 

 red. They are not yellow or bluish at any stage. The stamens and pistils are 

 also nearly white, the latter with rows cf oil glands showing as small grayish 

 dots. 



