STERILITIES OF WILD AND CULTIVATED POTATOES. 5 



that often many of the grains which swell and become plump may 

 contain very little granular material. These tests alone show that 

 the relative number of pollen grains aborted is high for many varieties 

 and that for others the pollen is evidently unable to function at all. 

 The method of culture utilized gave excellent germination, affording 

 a reliable test for the relative viability of pollen. 



Extensive tests were made to determine the most favorable medium 

 for testing the viability of pollen. Parallel series of cultures were 

 made with the following media: (a) Cane sugar in percentages of 

 5, 7i, 10, 12i, 15, 17i, 20, 22^, 25, 27^, and 30 in both tap and dis- 

 tilled water; (b) dextrin in the same percentages; (c) the addition 

 of 1 per cent agar to both a and h; (d) the addition to a, h, and c of 

 the extract of pistils made by grinding 10 pistils in 1 cubic centimeter 

 of water and filtering. In 1921 duplicate series involving a total of 

 several thousand tests were made for all these media, using pollen of 

 varieties and species which had previously given the best results. 



No germination was obtained with the use of any liquid medium. 

 The 7 per cent sugar solution reported by East (4) as giving germina- 

 tion was also used in every set of tests made in 1921 for 45 varieties. 

 The good pollen grains would swell to rotundity and small protrusions 

 appear at the various pores quite as shown by East (4) , but in this, 

 as in all other liquid media, no real germination was observed. 

 Excellent germination was obtained, however, with 1 gram agar and 

 15 grams cane sugar in 100 cubic centimeters of distilled water. 

 Occasionally 10, 12^, 17|, and 20 per cent sugar with 1 per cent agar 

 gave nearly as good germination. The sugar-agar media were made 

 up in bulk, run into test tubes, and sterilized. A fresh tube of each 

 grade was used each day. About one-fifth of a cubic centimeter of 

 a warmed and liquefied medium was placed on a glass slide so that it 

 spread out into a flat-topped circular drop about 1^ centimeters in 

 diameter and solidified. Pollen from dehiscing anthers was sprinkled 

 on the surface and the slide placed in a moist chamber at room tem- 

 peratures. During the season of 1922 the tests were made with the 

 10-f-l, 15+1, and 20 + 1 sugar-agar media, and the moist chambers 

 were kept overnight in a constant-temperature chamber set at a 

 minimum of 20° C. Final judgment of germination was made after 

 24 hours. 



Plates II to VI, inclusive, are photomicrographs of cultures on 1 

 per cent agar+ 15 per cent sugar kept in a moist chamber 24 hours 

 and then stained with acetocarmine. The imperfect grains, con- 

 taining little or no granular material, did not take the stain. 



ANTHERS AND POLLEN OF SUPPOSEDLY WILD SPECIES. 



Solanum chacoense. — The anthers of iS'. chacoense are plump, with 

 well-rounded and proportionately large pollen chambers (PI. I, 

 fig. 1). At least 99 per cent of the pollen from plants which flowered 

 during the period of the studies was plump and filled with granular 

 material. On 25 per cent sugar and 1 per cent agar only short pollen 

 tubes formed from about 5 per cent of the pollen; on 20 + 1 medium 

 about 10 per cent of the grains germinated; on the 15+1 medium 

 more than 90 per cent of the pollen grains germinated, the length of 

 many tubes being 800 m while some were 1,050 /z in length; on 10 

 + 1 medium about 20 per cent of the pollen germinated; on 7^ + 1 



