HORTICULTURE AND LANDSCAPE GARDENING. 71 



two and a half to four times greater. The very wide variations in the aver- 

 age numbers of seeds to a fruit, as indicated in the third column of figures, 

 is proof of the second point, i. e., that seed production bears no constant 

 ratio to the degree of development of fruit. 



It was observed that the seeds from Prussia germinated sooner than those 

 of American growth. This is, no doubt, due to the fact that tomatoes are 

 largely eaten in Germany in a green condition and that seeds are probably 

 saved from green fruits. It is well known that seeds from unripe fruits 

 germinate sooner than do those from ripe fruits. The percentages of germi- 

 nation are low in the German seeds. These facts do not always appear 

 prominently in the synoptical table, as some seeds received from Prussia 

 were not grown there. 



Training. — The culture of the tomato, being for the most part well under- 

 stood, is not a leading feature in this experiment. However, attention has 

 been given to the best methods. The area in plants was so large this year 

 that no training was given. The plants were allowed to fall upon the ground 

 freely. In a market patch leafless brush was laid alongside the plants and 

 it furnished a considerable support. I am inclined to regard it as desirable 

 in clean ground. The remainder of this paragraph is taken from the tomato 

 bulletin issued last year. Three methods of training were employed this 

 year, and another method last year. All the experimental tomatoes were 

 tied to one, two, or three stakes about four feet high. This method has 

 many disadvantages. It requires much labor to tie the plants, a labor which 

 must be repeated at short intervals throughout the growing season. The 

 tomatoes do not ripen evenly and early and it requires extraordinary time 

 and labor to pick them from the dense mass of stems and foliage. Although 

 the outlay for the stakes is small, this method of training is still the most 

 expensive of the four. La^t year we laid old boards lengthwise the rows and 

 close to the plants, supporting them upon pieces of scantling or blocks laid 

 upon the ground, and placed straw upon the boards. This method kept the 

 tomatoes clean, but it appeared to increase the rot. It certainly caused the 

 lower ripe tomatoes to rot prematurely. In our market patch this year we 

 adopted two sorts of racks. The first was a separate rack for each plant. A 

 stake was driven on either side of the plant, about fifteen inches from it, and 

 leaning so as to make an angle of about twenty degrees with the perpendicu- 

 lar. Upon these stakes three cross-slats were nailed, in the manner of a lad- 

 der. The plant was allowed to lop upon the racks. It was found necessary 

 to tie it, however, and even then branches slipped oif or broke themselves 

 over the slats. The second of these racks was continuous throughout the 

 row. About every six or eight feet a stout stake was driven on either side of 

 the row and fifteen inches from the plant, the stakes when firmly driven 

 standing some over a foot high. A strip of old boards was nailed near the 

 tops of the posts along either side of the row. Then edgings were tacked 

 across from one side to the other, four about each plant and a foot apart. 

 The plant now found itself growing up between the horizontal edgings, and 

 as it began to lop the rack held it above the ground. Upon this simple rack 

 the tomatoes needed no tying nor training and they spread themselves freely 

 to the sunlight. The circulation of the air under the racks was so free that 

 there was no unusual danger from rot. This is decidedly the best rack which 

 we have tried. We noticed, also, that the fruit ripened more uniformly here 

 than on the plants which were tied to stakes. 



