ing the anthers contain no pollen-grains and, indeed, do not 

 dehisce. 



The disintegration of the embryo-sacs is accomplished at a 

 somewhat later period in their history than that at which it takes 

 place in the pollen-sacs, but it also is complete before the time 

 of flowering. (Occasionally, however, a perfect egg-cell is pro- 

 duced, and it has been shown to be capable of fertilization by 

 the pollen from one of the seed-bearing varieties of the orange. 

 Such a X^avel, thus pollinated, will produce one or more seeds. 

 Experiment thus proves the correctness of the theory, empiric- 

 ally held by orange growers, that the presence of seed in a 

 navel fruit is evidence that the tree on which it was borne was 

 so situated as to permit fertilization from a seed sort. But as 

 the orange flower is insect fertilized, and the orchards are much 

 frequented by bees, the potent pollen may be carried from a 

 distance of a mile or more. 



Having ascertained the cytological causes which prevent the 

 production of seeds in the navel orange, we are brought to the 

 further inquiry as to how it acquired the seedless habit. But as 

 the conditions under which the first tree or trees originated — 

 for such a fruit may have originated more than once, — are not 

 known, this question cannot be definitely answered. Seedless- 

 ness in various plants has been found to result from a number 

 of different causes, among them hybridization and long con- 

 tinued intensive cultivation. In the absence of direct knowledge 

 these may be regarded as the probable originating causes which 

 led to the production of races of oranges, the Xavel among them, 

 in which the normal seed-bearing habit has been lost. 



As the navel orange is incapable of seed propagation, it is 

 evident that it can have no bearing upon questions of heredity ; 

 for heredity concerns itself with the transmission of qualities 

 from parent to offspring. The navel orange is multiplied and 

 distributed only by vegetative processes, and these artificial and 

 dependent upon the mechanical agency of man ; namely, budding 

 or grafting. Every Xavel tree, so far as the parts which grow 

 from the bud or graft are concerned, is just as much a part of 

 the parent tree as if it were a branch of it. They, therefore, 

 possess the qualities of the parent, not by inheritance, but direct- 

 Iv by physical identity, with it. Anv essential modification of 

 character could come onlv from bud-variation, which may occur 

 in any bud. but which, should it occur, can be propagated only 

 vegetatively. Since the occasional seed produced by a Navel is 

 the result of cross-fertilization, a tree grown from such a seed 

 would not be a navel orange, but a hvbrid between the Navel 

 and the staminate parent which contributed the pollen. In a 

 tree of this ancestry characters inherited from either parent 

 might appear. 



The second abnormalitv is the "navel." which gives this 

 orange its name. This consists of a small circular orifice at the 



47 



