640 ANNUAL REPORT OF THE Oft". Doc. 



has not been produced by a good plant we cannot expect it to 

 produce anylliinjj other than what has been bred into it. It is doubt- 

 less true that it is impossible to discriminate between good and poor 

 seed by looking at it, but by growing the crop we can readily see 

 important dillerences. l)uring the past lour years the Department 

 of Horticulture of the l*ennsylvauia Slate College has been making a 

 study of the importance of seed in the proti table production of cab- 

 bage and tomatoes. The work includes both variety and strain tests. 

 in the stiain tests the seeds of certain varieties are secured from 

 a large number uf seedsmen in various parts of the country. A ger- 

 mination test is made soon after the seed is received in order that 

 we may regulate the thickness of sowing. In a lot of seeds of one 

 variety secured in the spring of lUOl), the germination ranged from 2 

 to 99 per cent. A high percentage of geimination is not especially 

 important, yet it should not fall below 05 for cabbage and in any event 

 it is highly desirable that the percentage of germination be known 

 because of the reason just stated. 



After the germination test has been made the strains of the va- 

 riety to be tested are sown in Hats in the green-house. As germi- 

 nation proceeds, notable differences will be seen in the relative vigor 

 of the various strains as will be seen by some of the slides which 

 follow. 



This slide shows a flat of well grown plants ready for the field 

 planting. By careful control of temperature and watering vigorous 

 stocky plants may be grown. These are much superior to the leggy, 

 weak plants which develop where the temperature is too high and 

 watering too frequent. While the plants are in cold fiames they 

 are gradually accustomed to cold temperature so that in case severe 

 weather follows the field planting the plants will not be injured. 



During the past three years we have been making a variety test 

 of early varieties. These are of interest and value since some- 

 times new vai-ieties of merit are produced. At the present time we 

 have fifty so-called early varieties in the test but it is doubtful if 

 there is sufficient difference between many of the varieties to identify 

 them. 



In a slide which follows the variety Early Race Horse is shown 

 in comparison wdth a good strain of Jersey Wakefield. From the 

 illustration it will be seen that Race Horse matures several days 

 earlier than Jersey Wakefield. A test of other varieties shows that 

 Race Horse, Extra Early and First Early are identical. The tests 

 have shown that in several instances a well known variety has been 

 sent out under a new name. 



The next slide shows a field of ^'olga. This is a rather new va- 

 riety which appears to be well suited to a limestone soil. It is mid- 

 season in time of maturity. The heads are round, solid, and desirable, 

 except that the head leaves do not fold across as much as might be 

 desired. 



As previously stated, the strain tests are perhaps the most impor- 

 tant of the experiments we have been conducting during the past 

 three years. In a strain test of Jersey Wakefield secured in 1908, 

 which included twenty-five strains, notable differences were obseived 

 in the unif( Tmity to type, and the time of maturing. Of the twenty- 

 five strains u the test several matured more than 90 per cent, of the 

 crop within thirteen weeks of the time it was planted in the field. 



