824 
Telopea Vol. 6(4): 1996 
Using morphological and allozyme data, scored as both continuous and binary 
variables. Hopper & Burgman (1983) carried out phenetic and cladistic analyses of 
n populations representing both subspecies of E. caesia, with E. crucis subsp. lanceolata 
as the outgroup for the cladistic analyses. Their best estimate of phytogeny, derived 
by combining allozyme frequencies and morphometric data (Hopper & Burgman 
1983: fig. 3), showed subsp. magm to be monophyletic and nested deeply within 
subsp. caesia (Fig. 4). Subsp. caesia is paraphyletic, because subsp. magm is derived 
from an ancestral population of subsp. caesia (Hopper & Burgman 1983: 47). Therefore, 
the distinguishing features listed above should be interpreted as autapomorphies 
for subsp. magm. The authors of this study appeared quite comfortable with their 
conclusion that a recognised taxon (£. caesia subsp. caesia) was shown to be 
paraphyletic, and did not suggest any taxonomic rearrangement. 
Eucalyptus baxteri group 
The stringybark eucalypts are a monophyletic group restricted to mainland eastern 
Australia and characterised by an apomorphic bark type as well as distinctive hairs 
radiating from raised oil glands on the seedling leaves (Brooker & Kleinig 1983; Ladiges 
& Humphries 1986). In a cladistic analysis of the group using morphology of seedlings 
and adults, Ladiges and Humphries (1986) found a clade comprising £. deuaensis, 
E. baxteri and £. alpim. For the purpose of analysis, they recognised two forms within 
£. baxteri (South Australian and Victorian) and two within £. alpim (Mirranatwa Gap 
and Victoria Range). In their phylogeny, £. alpim was monophyletic but £. baxteri was 
paraphyletic, with the South Australian form sister-taxon to the Victorian form plus 
£. alpim. Morphometric studies on populations of £. baxteri (Marginson & Ladiges 
1988) found two distinct phenetic clusters corresponding with the above geographic 
forms, which were treated as two species: £. baxteri sens. str. and £. aremcea, sp. nov. 
However, no morphological autapomorphy was discovered for either segregate species. 
Subsequently, Whiffin & Ladiges (1992) investigated variation in leaf volatile oils among 
populations of £. aretiacea, E. baxteri sens. str. and £. alpim, using phenetics and dadistics 
based on distance data (leaf volatile oil composition cannot be expressed as discrete 
quantitative characters). Their sampling of populations of £. alpina was more 
comprehensive than in the previous studies. Although they obtained different 
phytogenies depending upon the tree-building method, all agreed that £. areriacea is 
monophyletic, whilst both £. baxteri sens. str. and £. alpim are paraphyletic (Fig. 5). 
Eucalyptus alpim was found to comprise three distinct (autapomorphic?), allopatric 
forms appearing to originate independently, either from within £. baxteri, or sharing a 
common ancestor with £. baxteri (Fig. 5). Each was described as a species (Ladiges & 
Whiffin 1993). This pattern, consisting of several distinctive, apparently autapomorphic 
species nested within a widespread, variable paraphyletic species is very similar to 
that shown by Daviesia ulicifolia (above). These authors did not explicitly discuss the 
consequences of knowingly circumscribing £. baxteri as a paraphyletic species, although 
to be fair they did not change its status —■ it had always been paraphyletic. 
Corallorhiza maculata 
In a detailed and comprehensive study, Freudenstein & Doyle (1994) constructed a 
well-corroborated phylogeny of 35 populations of Corallorhiza maculata (Orchidaceae) 
and two closely related species using restriction fragment variation of plastid DNA. 
All three species were recognised under the phylogenetic species concept because 
they were diagnosable by morphological characters. Plastome types corroborated 
these diagnoses. The plastid phylogeny showed C. maculata to be paraphyletic because 
both other species were nested within it. Acknowledging this, the authors were 
nevertheless prepared to recognise all three species. 
