322 
differing indumentum patterns. None of these 
varieties was accepted by Australian botanists 
of the time. 
Keng (1969) used the name Anisomeles 
salviifolia for some Malesian material, but 
later he (Keng 1978) considered that even A. 
salviifolia could not be maintained, reducing 
it to a synonym of A. malabarica (L.) R.Br. 
On this basis, the name A. malabarica was 
subsequently adopted by all Australian 
herbaria for nearly all Australian Anisomeles 
specimens, and this has been mainly the 
status quo until this revision. Bhatti & 
Ingrouille (1996) used the name A. salviifolia 
for specimens originating in Australia and 
New Guinea, although this was partly based 
on the erroneous belief that all four anthers 
in A. salviifolia are dithecous. In fact, all 
Anisomeles species have the same pattern of 
two dithecous anthers and two monothecous 
anthers in each flower. 
Harley et al. (2004) accepted three 
species for the genus, but did not name 
them. Govaerts et al. (2013) have accepted 
five Anisomeles species, with four occurring 
in southern Asia and Malesia (A. candicans 
Benth., A. heyneana Benth., A. indica (L.) 
Kuntze, A. malabarica ), and two occurring 
in Australia (A. malabarica, A. salviifolia). 
In the current paper, the four Asian/Malesian 
species accepted by Govaerts et al. (2013) are 
maintained, and two are added; A. principis 
A.R.Bean (found in Timor) and A. papuana 
A.R.Bean (found inNewGuinea, the Moluccas 
and Torres Strait, Queensland). For Australia, 
23 species are recognised, being A. salviifolia 
and 22 other species, including 18 species 
newly named. The present study clearly 
demonstrates that A. malabarica does not 
occur in Australia. The taxonomic Tumping’ 
currently applied to Australian Anisomeles is 
a legacy of the statements of Bentham (1870), 
reinforced by Keng (1978), to the effect 
that Anisomeles is a diffuse genus, where 
intergradation is rife and there are no discrete 
taxa. The current author has found that this is 
not the case. While there is certainly evidence 
of intergradation between some species, 
many are eminently discrete and separable 
from each other by discontinuities in readily 
Austrobaileya 9(3): 321-381 (2015) 
observable morphological characters. The 
differences are often subtle, but they are at 
least as robust as differences used to separate 
species in other genera of Lamiaceae, and in 
other related families. 
Rothmaler (1944) maintained that 
the generic name Epimeredi Adans. is 
synonymous with Anisomeles , and that 
the former should replace the latter. As a 
consequence, Epimeredi was taken up for 
a time by some Australian herbaria. An 
unsuccessful proposal to conserve Anisomeles 
against Epimeredi was made by Subramanyam 
& Henry (1969). In the meantime, Backer 
and Backhuisen van den Brink (1965) had 
decided that Adanson’s original material 
of Epimeredi probably did not include any 
specimen referable to Anisomeles , and so 
they maintained Anisomeles as the accepted 
generic name. Parkinson (1987) considered 
that none of Adanson’s unconserved generic 
names were validly published. However, this 
conclusion was refuted by Wilbur (1989), and 
the current consensus amongst the botanical 
community is that Adanson’s genera are 
validly published. 
I have examined high quality images of 
the four Lamiaceae specimens in the Adanson 
Herbarium (P-Ad.) that have Epimeredi 
written on the label; these four specimens 
comprise three different species. Only one of 
these specimens (P00680377) can be linked 
to the protologue, as one of its labels includes 
the notation “h. Reg Par 1754”. This is a 
reference to the Hortus Regius Parisiensis , 
cited in the protologue. The date 1754 
precedes the publication date for Epimeredi , 
and the specimen is therefore original 
material. This specimen is readily identifiable 
as Anisomeles malabarica. The inevitable 
conclusion is that Epimeredi Adans. (1763) 
and Anisomeles R.Br. (1810) are congeneric. 
A proposal to conserve Anisomeles against 
Epimeredi is currently under consideration 
by the Nomenclature Committee for Vascular 
Plants (Bean 2015). 
Materials and methods 
This revision is based mainly on a 
morphological examination of 1300 herbarium 
