147 



State, sweet Oranges are reported to be grown wholly or largely from 

 seeds in the following places : Morocco, Guerrero (Mexico), Ecuador, Ja- 

 maica, Guadeloupe, Porto Eico, Syria, Sidon, Philippine Islands, Naples, 

 Azores Islands, Spain, &c* 



Far the greater portion of the evidence, it will be seen from the above 

 discussion, supports the belief that seeds from sweet Oranges commonly 

 produce sweet fruits. When bitter or sour fruits result from planting 

 sweet Orange seeds it is probable that other seeds have been acci- 

 dentally mixed with them or that they are the results of accidental 

 hybridisation. Sweet Oranges are almost invariably grown near sour 

 Oranges, Lemons, Limes, Pomelos, &c, and it is not greatly to be won- 

 dered at that from seeds formed under such conditions a hybrid occa- 

 sionally arises Both Fawcett and Syme think this is what misled Mac- 

 fadyen. Stubbs and Morganf also mention this as one of the main 

 causes of the variation in the quantity of seedling Oranges. That 

 perfectly fertile seeds may be obtained by hybridising the different spe- 

 cies of Citrus I have incidentally proved in the course of pollination ex- 

 periments which I have been conducting under the direction of the U.S. 

 Department of Agriculture. The following instances may be cited : — 



Parson Navel Orange (a local variety of the common sweet Orange, 

 Citrus aurantium), which is commonly seedless, crossed with pollen from 

 the acid Lime (Citrus limetta), produced three full seeds, which were 

 planted, and gave four seedlings, two embryos of one seed developing. 



St. Michael Blood Orange (Citrus aurantium), crossed with pollen of 

 Grape-fruit or Pomelo (Citrus decumana), formed fifteen well developed 

 seeds, which were planted, and gave twenty-two seedlings. If seeds 

 thus artificially hybridised are fertile, there is no reason to d mbt that 

 seeds accidentally hybridised may be fertile. 



In each of the above cases the number of seedlings resulting is in ex- 

 cess of the number of seeds planted This introduces an interesting 

 feature into experiments of this nature. It is a well-known fact that 

 most species of citrus fruits produce several embryos, some seeds pro- 

 ducing as many as twenty-five ; only from one to three of thes % how- 

 ever, commonly develop. Cruder, Schacht, Hofmeister, Strasburger, 

 and others have carefully studied this interesting polyembryouic de- 

 velopment ; and it appears, so far as we yet know, that only one of the 

 embryos in each seed is formed by the act of fecundation, the others 

 being what are termed adventive embryos developing from the side of 

 the embryo sac. It is thus probable that only the sexually-developed 

 embryo will be influenced by the hybridisation. As two to three embryos 

 frequently develop from a seed, even in cases of hybridisation some of 

 the seedlings, those from adventive embryos, will probably produce true 

 to the parent bearing the fruit. 



While seedlings of the sweet Orange produce sweet fruit similar to 

 those produced by the parent, there is yet much variation in the fruit 

 from such seedlings. Commonly it can hardly be distinguished from 

 the parent, but many are of much inferior quality, and occasionally one 



* Fruit Culture in Foreign Countries. Special Consular Reports, Washington 

 1890. 



t Stubbs and Morgan, The Orange and other Citrus Fruits from Seed to Market. 

 Special Bulletin, Louisiana State Experiment Station, 1883, p. 12. 



