357] 358 ALISO [VoL. 5, No. 3 Species in several other subgenera were studied as follows: In the subgenus Anogra five species were included: 0. deltoides in the Colorado and Mohave deserts of California, 0. wigginsii in Baja California, 0. californica in the Mohave desert of California, 0. engelmannii in Texas, and 0. runcinata in New Mexico. In the subgenus Raimannia three species were studied: 0. albicaulis in Texas, 0. drummondii along the Gulf coast of Texas, and 0. laciniata in central

more »

... s. In the subgenus Pachlophis two subspecies of 0. caespitosa were examined at several stations from California to Arizona and western Colorado. Two species of subgenus Galpinsia were studied: 0. greggii and 0. hartwegii, both in western Texas. Of the subgenus Lavauxia one species, 0. taraxacoides, was studied in the Sacramento Mountains of New Mexico. A single species in subgenus Megapterium, 0. brachycarpa, was studied in the Charleston Mountains near Las Vegas, Nevada. These species were studied in from one to five localities but for only a single night and morning at each, so that the data are not as complete as for 0. hookeri. These subgenera show marked variation in floral morphology, although definitely on a common theme and always exhibiting the complex of characteristics listed above. Represented among them are species with quite different breeding systems. This study has yielded evidence on the following problems. (1) The relative importance of hawkmoths as pollinators of the species of Oenothera studied and the exact description of such pollination. (2) The question of whether morphological differences in the flowers adapt them to specific pollinators and, if so, if pollination is restricted to these particular animals. ( 3) The change in spectrum of pollinators for a single species or morphological type of Oenothera flower with changes in geographical area or in habitat. ( 4) The occurrence of differences in breeding system between species of Oenothera and the significance of such differences. ( S) The abundance and behavior of the pollinating insects. METHODS The main method used during this research was direct field observation. This was supplemented in several ways. Photographs of the flowers and of the visiting animals were taken with a 3S mm camera. A strobe light was used for photography after dark. Moths and other insects were captured for identification, to measure morphological characteristics having to do with pollination, and for a capture-recapture experiment. They were taken either at the flowers with a net or at a light-trap composed of a 1S-watt ultraviolet lamp set up in front of an ordinary sheet. This trap was usually set up within 10 yards of the plants. At most of the colonies studied, a series of SO flowers was emasculated about dusk and checked during the night and following morning with a series of SO unemasculated flowers. This technique is described in more detail later. Vouchers of the plants studied were collected and have been deposited in the herbarium of the Rancho Santa Ana Botanic Garden. Flowers were often collected in FAA for study in the laboratory as well. Specimens of a number of species were grown at the Rancho Santa Ana Botanic Garden for determination of self-compatibility relationships. This work was done in cooperation with Mr. William Klein.