780 
Telopea Vol. 6(4): 1996 
Table 2. Flowering phenology of Persoonia species from which bees were collected for this study; 
each entry is the number of flowering specimens collected during a calendar month, held by 
NSW; sets of multiple collections made at the same site on the same day were each treated as 
a single record. 
Feb Mar Apr May Jun Jul Aug Sep 
6552---- 
- 2 - - 1 - - - 
15 4 - 1 - - 1 - 
acerosa 
arborea 
asperula 
chamaepeuce 
chamaepitys 
glaucescens 
isophylla 
lanceolata 
laurina subsp. laurina 
levis 
microphylla 
mollis 
myrtilloldes 
subsp. myrtilloldes 
nutans 
oblongata 
oxycoccoides 
pinifolia 
silvatica 
subvelutina 
virgata 
Oct Nov Dec Jan 
- - 1 9 
- - - 6 
1 - 6 25 
- 6 11 2 
- - - 2 
- - 1 1 
2-24 
15 9 3 
2 5 8 2 
- - 4 1 
2 1 8 18 
- - 10 9 
- 1 7 4 
- - - 2 
- - 3 5 
- - 3 8 
1 1 4 11 
- - - 3 
1 1 3 11 
4 1-1 
2 3 3 - 
8 7 6 5 
1 1 - - 
3 3 12 
1 1 - - 
21 21 12 9 
4 2 11 
- 1 1 - 
4 3 2 1 
111- 
6 3 11 
1 - - - 
7 1 - - 
3 3 14 
1 
3 - 2 - 
- - - 2 
4 5 4 3 
2 3-1 
1 _ _ _ 
1 - - - 
2 2 - - 
1 - 1 - 
For each taxon in this study, one tepal (plus its attached stamen) was removed from 
most, or all, of the flowering specimens held at NSW. Care was taken not to sample 
from multiple duplicates of a single collection. The tepal/stamens were rehydrated 
by soaking in distilled water overnight, which effectively restored their fresh, three- 
dimensional form. For each tepal/stamen, the length of the floral tube was measured 
under lOx magnification, using an eyepiece micrometer fitted to a stereo dissecting 
microscope. The length of the floral tube was defined as the linear distance between 
the base of the staminal filament (where it is adnate to its tepal) and the point at 
which the stamen becomes free from its tepal (at, or slightly above or below, the base 
of the anther). In most species, this is the point at which the tepal starts to recurve, 
thus losing coherence with the adjacent tepals. Measurements were scored by taxon 
and summary statistics (sample size, mean, range, standard deviation) calculated. 
Sampling floral odour follows Bernhardt (1995) and Buchmann et al. (1978). Fresh 
flowers were placed in clean, glass vials and sealed for 15 minutes, 30 minutes, 
45 minutes, one hour and two hours. The vials were placed in a warm, sunny 
location, then reopened and smelled. To determine possible sites of scent glands 
(osmophoric activity) whole flowers of seven species were submerged in a 1 % solution 
of Neutral Red for two to 24 hours, then washed in distilled water for two hours. 
Living flowers of P. cornifolia (living collections number 973375) and P. katerae (living 
collections number 877128) used for scent tests came from shrubs grown at the 
Mount Annan Botanic Garden and the Royal Botanic Gardens, Sydney, repectively. 
