The Laevifolia-F amily . 267 



Near the field, there was a small heel in a park in 

 which some annuals were grown every year. Amongst 

 these was Oenothera Lamarckiana which spread from 

 this spot over the field. When I first visited the place 

 the little bed had long since run wild, but was still recog- 

 nizable. The Oenotheras were most numerous on the 

 northeast corner of the field, close to the bed they spread 

 from. Here they formed a dense jungle of branched 

 stems as high as a man. The area of the whole field vv^as 

 about 5000 square meters. 



This dense group was surrounded by a broad zone 

 of plants a few^ of which were in flower, and of numerous 

 rosettes, growing together; whilst to the west of this 

 * zone only isolated rosettes were to be found. But on the 

 greater part of the field not a single Oenothera was to be 

 seen at first glance ; though, after further search I found 

 flowering plants in two places, and in five or six other 

 places rosettes with radical leaves, evidently from seed 

 which had germinated that spring. 



We may conclude therefore that the plants began to 

 spread from the northeast corner of the field in the period 

 1884 to 1886. In 1888 the whole field had become occu- 

 pied by isolated groups consisting of both young and 

 two-vear-old flowerins^ individuals. 



In the winter 1886-7 and 1887-8 a part of the north- 

 east corner of the field comprising more than one-half 

 of the Oenothera patch was deeply dug and planted with 

 oaks. But the ground was so full of seed that for two 

 summers afterwards, this piece was thickly covered with 

 rosettes and flowering plants of Oenothera. One result 

 of this was a very high degree of variability in the dura- 

 tion of life of the plants, for there could be seen on this 

 spot during the month of August and September numer- 



