(Arft 1995a). Much more work is needed to quantify these variables in the 

 Colorado and Utah sites, or apply them in the rest of its range. 



The ten known occurrences of Spiranthes diluvialis in Montana total over 

 1,400 flowering plants observed. Survey of plant numbers was incomplete or 

 non-existent at two sites due to access restrictions but was comprehensive at 

 small sites. The numbers of flowering plants in the two largest populations 

 had a tally of over 500 plants in 1997, a favorable year for high flowering 

 levels. Only one plant could be found in the smallest population. 



These one-time counts are hampered by three factors, so that flowering stem 

 coimts are to be taken as very conservative low estimates of population size. 

 First, mature plants may be in three alternate states, only one of which is 

 easily located. These are the flowering state, the vegetative state with basal 

 leaves and the seasonal dormancy state. The vegetative basal leaves are 

 shorter than the canopy of grasses, and the narrow, linear leaves are difficult 

 to discern among grass leaves (see photographs of plants as vegetative basal 

 leaves in Appendix B). The dormancy state has no above-ground growth. 



The transition between states has been shown to fluctuate significantly. At the 

 City of Boulder site, flowering stem counts have varied fi-om 5,435 stems in 

 1986 to 131 stems in 1998 (U.S. Fish &, Wildlife Service 1992). Subsequent 

 monitoring has determined that the apparent decline is primarily attributable 

 to the non-flowering state changes rather than mortality (Arft 1995a). 

 Flowering stem counts were made at one Montana site (EO#001) in four 

 consecutive years, in a population estimated to have over 1 00 plants, but with 

 annual flowering stem numbers ranging from 26-71 between years. 



Second, the emergence of flowering shoots can be staggered by about four 

 weeks, with some plants flowering in early August and fruiting by the end of 

 the month, while a few plants are just beginning to flower in September (pers. 

 obs. ). Even a count in peak flowering is apt to miss at least a trace of 

 population numbers. 



The third factor hinges on fieldwork resources. The species habitat cannot 

 dependably be located without aerial photos to show the wetlands. It is 

 difficult for an investigator to gaze across a seemingly level plain and pick 

 out all of the channels less than 1 m deep snaking across it, particularly 

 channels lacking shrub borders or having high density. The set of over 300 . 

 aerial photo xeroxes that were carried into the field proved to be incomplete, 

 and four of the new occurrences were surveyed without benefit of aerials so 

 the surveys are to be considered incomplete. 



30 



