128 
Fig. la. Dactylospongia elegans (Thiele). USNM 
23707. 
subdermal cavities, bet. abundant over most of 
the surface. 
SKELETON: A compact irregular reticulation 
of clear yellow-brown fibres; many ascending 
fibres are distinguishable only immediately be- 
low the surface. Some stratification is shown in 
most fibres and a poorly demarcated pith occurs 
infrequently. Isolated spicule fragments occur 
randomly. A characteristic feature of the skele- 
ton is the granulate surface of the fibres. Fibre 
diameter is 20-54/*, (38g). 
FLAGELLATE CHAMBERS: Small, spherical, 
24-30 fx in diameter, and are regularly distrib- 
uted throughout the body of the sponge except 
in the lacunar areas which surround the sub- 
dermal cavities. The endosome contains many 
darkly pigmented granules and abundant dia- 
tom skeletons. 
DISCUSSION: Thiele (1899) described this 
sponge from the Celebes and established the 
genus Luffariella to receive Luffaria variabilis 
Polejaeff and Luffariella elegans; the former was 
designated as the type of the genus by Thiele 
and later by de Laubenfels (1936). 
PACIFIC SCIENCE, Vol. XIX, April 1965 
Considerable confusion has centered around 
the correct generic position of Luffaria vari- 
abilis. The genus Luffaria was unrecognizably 
described by Duchassaing and Michelotti (1864). 
Schmidt (1870) redescribed the genus with- 
out adding any species, and Polejaeff (1884) 
described Luffaria variabilis. Lendenfeld (1889) 
recognized that Polejaeffs sponge was not a 
Luffaria as understood by Schmidt ( = Veron - 
gia) , and took the erroneous step of re-estab- 
lishing Luffaria under Polejaeffs name to re- 
ceive L. variabilis and two new species. Topsent 
(1934) considered that both Luffaria and Luf- 
fariella should be abandoned and all species in- 
volved transferred to Aplysinopsis Lendenfeld. 
The difference between Aplysinopsis and Tho- 
recta Lendenfeld is very slight, merely that 
Thorecta has a relatively smooth surface and 
more foreign material in the dermis ( see de 
Laubenfels, 1948). These two genera are now 
considered synonymous. 
The type specimen of Luffaria variabilis 
Polejaeff should be established as BM 85.8.8.52. 
This sponge has been examined and in most 
features it is compatible with Cacospongia. It 
has, however, one distinguishing characteristic: 
an extremely fine and regular tertiary fibre net- ! 
work is present throughout the sponge. In 
view of this marked difference in the skele- 
ton, it is proposed to retain Luffariella for L. 
variabilis and other sponges which have com- < 
parable skeletal structure. Luffariella geometrica | 
Kirkpatrick may properly belong here. 
It is clear that Luffariella elegans Thiele is 
genetically distinct from Luffariella variabilis. \ 
Examination of the Palau specimen of L. j 
elegans and an undescribed specimen in the 
British Museum (1946.11.25.170) shows that 
the correct generic grouping for this sponge is 
not near Cacospongia but near Hippospongia, 
differing from some specimens of this genus 
only in form, the presence of long surface chan- 1 
nels, and the lack of any cored primary fibres. 
These are the only features separating L. ele- 
gans and Hippospongia metachromia 2 (de Lau- 
benfels), also from the Palau Islands. 
2 No cored ascending fibres were described for this j 
sponge, but re-examination of the type has shown 
them to be present in the subdermal region. 
