112 THE PLANT WORLD 



Regarding the reproduction of the bush morning-glory, it may be 

 generally concluded that most of the plants grow from seeds. The num- 

 ber of seeds produced is very large, a mature plant bearing a thousand or 

 more blossoms each season, each of these, if fully pollinated, producing 

 four seeds. It has not been the fortune of the writer to collect seedlings 

 in the field, but through the courtesy of Mr. Theodor Holm, he has seen 

 a seedling raised in Washington, from seeds from Kansas. The striking 

 thing about the seedling is the great similarity between the cotyledons 

 or seed leaves of this species and those of the common climbing morning- 

 glory i^Ipomoea purpurea) . 



The foregoing facts make emphatic the finding, in Grant County, 

 Nebraska, July 17, 1903,* evidence of another method of reproduction of 

 this species. This method is vegetative, and consists of the formation of 

 new plants at the apex of horizontal, or nearly horizontal, lateral root- 

 shoots (see Plate VI). The writer, with three Nebraskans, had dug out 

 one of the large plants to ascertain the size of the root. This was in 

 Cherry County, on the 15th of July. One of the ranchmen stated that 

 the plants start with roots not having an enlarged storage part. In Grant 

 County there were numerous young plants, so search was made for the 

 youngest plants, seedlings if possible. No seedlings were found, but a 

 tiny plant was noted which evidently was of the then present season. 

 It was slowly dug out, and was found to have, besides its vertical root, 

 a horizontal one extending from the upper part of the vertical root. This 

 was traced up the slope of the sand hill for fifteen or twenty deci- 

 meters, where it was found growing from the upper part of an older ver- 

 tical root. The parent root probably was five or more years old. Of 

 course the horizontal root-shoot then proved to be not a part of the 

 younger plant itself. This method of reproduction is quite exten- 

 sive among plants, but Mr. Holm informs me that only two other 

 species of the morning-glory family (^Convolvulaceae) have the same re- 

 corded for them. Irmisch described it for Convolvzdiis althaeoides , and 

 Warming for Convolvuhis arvensis, both of the Old World, and the second 

 naturalized in this country. 



One character of the root-shoot of the bush morning-glory deserves 

 notice. The bud terminating the root-shoot, and about to form a new 

 plant, appears to be formed backwards if that term may be used. In- 

 stead of pointing away from the apex of the root-shoot, the bud appar- 

 ently starts growth in the direction of the dorsal surface of the root- 

 shoot. The advantage of this is the rapid formation, at the proper time, 

 of disruptive tissue ; not at the base of the bud which is to produce the 

 downward vertical root of the new plant, but by the part which is to pro- 



* Appreciative acknow^ledgment for the above opportunity is hereby made to the Division of 

 Grass and Forage Plant Investigations, United States Department of Agriculture ; June-Sejitember, 

 1903- 



