(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Proceedings of the Linnean Society of New South Wales"

THE 



PROCEEDINGS 



OF THE 



LiMMEnn Society 



OF 



New South Wales 



FOR THE TEAR 

1923 

Vol. XLVIII. 




WITH FIFTY -TWO PLATES 

and 438 Text-figures. 



Sydney: 
PRINTED AND PUBLISHED FOR THE SOOXBTT BY 

THE SYDNEY AND MELBOURNE PUBLISHING CO., LTD. 
29 Alberta Street. Sydney. 

AND 

SOLD BY THE SOCIETY. 



1923-1924 




COfiTENTS OF rROCEEDlNGS, 1923 

PART I. (No. 190.) 
{Issi(,ed 13th April, 1923.) 

Pages. 
Presidential Address, delivered at the Forty-eighth Annual Meeting, 

28th March, 1923, by G. A. Waterhouse, B.Se., B.E., F.E.S. 

[Plates i.-ii.] i.-xxiii. 

Elections and Announcements , xsiv. 

Hon. Treasurer's Balance Sheets xxv.-xxvii. 



PART II. (No. 191.) 
{Issued 15th June, 1923.) 

The High Temperature Organism of Fermenting Tan-bark. Part ii. By 
R. Greig-Smith, D.Sc, Maeleay Bacteriologist to the Society. (Ten 
Text-figures) 1-16 

Studies in Australian Entomology. No. xviii. New Genera and Species 

of Carabidae. By Thomas G. Sloane 17-39 

New Termites from Central and South-east Australia. By Gerald F. 

Hill. (Twenty-one Text-figures) 40-48 

Studies in Life-histories of Australian Diptera Brachycera. Part i. 
Stratiomyiidae. No. 4. The respiratory system in larva, pupa and 
imago of Metoponia rubriceps Macquart. By Vera Irwin-Smith, 
B.Sc, F.L.S., Linnean Maeleay Fellow of the Society in Zoology. 
(Fifty Text-figures). 49-81 

Studies in Symbiosis, iii. Contribution to the Morphology and Physio- 
logy of the Root-nodules of Podocarpus spinulosa and P. elata. By 
J. McLuckie, M.A., D.Sc. (Twenty-one Text-figures) 82-93 

A Revision of the Australian Diptera belonging to the Genus Sarco- 
phaga. By Professor T. Harvey Johnston, M.A., D.Sc, and G. H. 
Hardy. (Twenty-eight Text-figures) 94-129 

The Lorcmthaceae of Australia. Part iv. By W. F. Blakely. (Plates 

iii.-xiv.) 130-152 

Some Notes on the Permo- Carboniferous and overlying Systems in 

Central Queensland. By H. I. Jensen, D.Sc 153-158 

Revision of the Genera Ethon, Cisseis and their Allies [Buprestidae] . 

By H. J. Carter, B.A., F.E.S. (Twelve Text-figures) 159-176 

The Life-history of Microeachrys tetragona (Hook.). By Professor A. 
Anstruther Lawson, D.Sc. (Plates xv.-xvi. and Thirty-four Text- 
figures) 177-193 

Studies in Symbiosis. iv. The Root-nodules of Casuarina Cunning- 
hamiana and their Physiological Significance. By J. McLuckie, 
M.A., D.Sc. (Sixteen Text-figures) 194-205 



r» 



IV. CONTENTS. 

PAET III. (No. 192.) 
(Issued 3rd October, 1923.) 

Notes on Freshwater Algae. By the late G. I. Playfair. (Thirty 

Test-figures) 206-228 

Studies in the Vegetation of Arid and Semi-arid New South Wales. 
Part i. The plant ecology of the Barrier district. By Marjorie I. 
CoUins, B.Sc, F.L.S., Linnean Macleay Fellow of the Society in 
Botany. (Plates xv.-xxiii. and six Text-figures) 229-266 

A Monograph of the Freshwater Entomostraca of New South Wales. 
Part iii. Ostracoda. By Marguerite Henry, B.Sc, Linnean Macleay 
Fellow of the Society in Zoology. (Plates xxiv.-xxix.) 267-286 

Anatomical Features of the Mature Sporophyte of Selaginella uliginosa. 
By Jessie K. Steel. (Nineteen Text-figures). (Communicated by 
Professor A. A. Lawson) ' 287-300 

Studies on Australian Mollusea. Part xiv. By C. Hedley. (Plates 

xxx.-xxxiii. and one Text-figure) 301-316 

A Eevision of the Australian Species of the Genus Bassia. By R. H. 

Anderson, B.Se. (Agr.). (Plates xxxiv.-xxxvi.) 317-355 

Studies in Plant Pigments, i. The yellow colouring matter of the 
Acacias. By J. M. Petrie, D.Sc, F.I.C., Linnean Macleay Fellow 
of the Society in Biochemistry. (Five Text-figures) 356-367 

Studies in Life-histories of Australian Diptera Brachyeera. ii. Asilidae. 
No. 1. Catalogue of the species of Asilidae of which the earlier 
stages have been recorded. By Vera Irwin-Smith, B.Sc, F.L.S., 
Linnean Macleay Fellow of the Society in Zoology 368-374 

Studies in Life-histories of Australian Diptera Brachyeera. ii. Asilidae. 
No. 2. Notes on the egg-laying, eggs and young larvae of Neoaratus 
liercules Wied. By Vera Irwin-Smith, B.Sc, F.L.S., Linnean Macleay 
Fellow of the Society in Zoology. (Eighteen Text-figures) 375-380 

Revision of the Amyeterides (Coleoptera). Part viii. The Euomides. 

By E. W. Ferguson, M.B., Ch.M 381-435 

Studies in Symbiosis, v. A contribution to the physiology of Gastrodia 
sesamoides (R.Br.). By J. McLuekie, M.A., D.Sc. (Sixteen Text- 
figures) 436-448 



PART IV. (No. 193.) 
(Issued 14th December, 1923.) 

A new Conifer from Southern Queensland. By C. T. White. (Plate 

xxxvii.) 449-450 

A Revision of the Australian Anerastrianae (Lepidoptera). By A. J. 

Turner, M.D., F.E.S 451-461 

On some Abnormal Sugar-canes. By T. Steel. (Plate xxxviii.) 462-464 

The Strophomenidae from the Fossiliferous Beds of Bowning, N.S.W. 

Part i. Stropheodonta. By John Mitchell. (Plates xxxix.-xlii.). 465-474 

The High Temperature Organism of Fermenting Tan-bark. Part iii. By 
R. Greig-Smith, D.Sc, Macleay Bacteriologist to the Society. (One 
Text-figure) 475-480 



CONTENTS. V. 

Mesozoie Insects of Queensland. No. 10. Summary of the Upper 
Triassie Insect Fauna of Ipswich, Q. (With an Appendix describ- 
ing new Hemiptera and Planipennia) . By R. J. Tillyaxd, M.A., 
Sc.D., D.Sc, C.M.Z.S., r.L.S., F.E.S. (Plate sliii., and four 
Text-figatres) 481-498 

The Life-history of PJierosphaera. By Professor A. Anstruther Lawson, 

D.Sc, F.R.S.E., F.L.S. (Plate xliv., and thirty-one Text-figures). 499-516 

Note on the Genus Synechocera, with Description of a New Species. By 
A. Thery. {Communicated by H. J. Carter, B.A., F.E.S.) . (One 
Text-figure) 517-518 

On some Australian Galerueides (Coleoptera, Chrysomelidae). By A. M. 

Lea, F.E.S. (Sixty Text-figures) 519-575 

Australian Neuroptera. Part iv. By P. Esben-Petersen. {Communicated 

hy W. W. Froggatt, F.L.S.) . (Plates xlv.-xlvi) 576-592 

Australian Neuroptera. Part v. By P. Esben-Petersen. {Com- . 

municated hy W. W. Froggatt, F.L.S.). (Plate xlvii.) 593-600 

Notes on Australian Diptera, with descriptions. By J. R. Malloeh, 

{Communicated hy E. W. Ferguson, M.B., Ch.M.) 601-622 

The High Temperature Organism of Fermenting Tan-bark. Part iv. The 
effect of chill. By R. Greig-Smith, D.Sc, Macleay Bacteriologist 
to the Society ". 623-633 

A Contribution to our Knowledge of the Fueaceae. By May M. Williams, 

B.Sc (Twenty-three Text-fig-ures) 634-646 

Fissicorn Tachinidae, with description of new forms from Australia and 
South America. By Professor M. Bezzi. {Communicated by E. W. 
Ferguson, M.B., Ch.M.). (Eight Text-figures) 647-659 

The Occurrence of secretory Canals in certain Myrtaceous Plants. By 
M. B. Welch, B.Sc, A.I.C. (Plates xlviii.-l., and three Text- 
figures) 660-673 

Preliminary Note on the embryo sac of Styphelia longifolia (R.Br.). By 

Patrick Brough, M.A., B.Sc, B.Sc. Ag. (Twelve Text-figures). . 674-680 

New or notew9rthy Plants from the National Herbarium, Sydney. By E. 

Cheel 681-688 

PART V. (No. 194.) 

{Issued 15th February, 1924.) 

■ Abstract of Proceedings xxix.-xxxviii. 

Donations and Exchanges xxxix.-liii. 

List of Members liv.-lviii. 

Index lix.-lxxxii. 



VI. CONTENTS. 

LIST OF NEW GENERIC NAMES PROPOSED IN THIS VOLUME (1923). 

Adeloplectron (Myrmeleonidae) . . 577 Myrmecodemus (Odaeanthini, Cara- 

A.ulacoliws (Odacanthini, Carabidae) 32 bidae) 33 

Cryptocladocera (Taehinidae) . . . 653 Osmylopsy chops (Prohemerobiidae) 496 
Dictyoleon (Myrmeleonidae) . . . . 584 Platyleon (M3rrmeleonidae) . . . . 578 
Euomella (Euomides, Coleoptera) . 388 S^tenogymnocnemia (Myrmeleoni- 
dae) 581 



CORRIGENDA. 

Through an error in numbering, the first two plates accompanying the paper 
by Marjorie I. Collins in Part 3 of the Proceedings were numbered xv. and xvi., 
which two numbers had already been given to the two plates accompanying the 
paper by Professor A. Anstruther Lawson in Part 2. 

Plate xlviii.^ — Reverse block of figure 4. 



LIST OF PLATES. 

PROCEEDINGS, 1923. 

i.-ii. — ^A'ariations of Tisiplione abeona Joanna. 

iii.-xiv. — Loranthaceae of New South Wales. 

xv.-xvi. — Microcachrys tetragona. 

XV. bis.-xxiii.- — Vegetation of the Barrier District, N.S.W. 

xxiv.-xxix. — Ostracoda of New South Wales. 

xxx.-xxxii. — Australian Mollusca. 

xxxiii. — Tethys angasi Sowerby. 

xxxiv.-xxxvi. — ^New Species of Bassia. 

xxxvii. — CalUtris Baileyi, n.sp. 

xxxviii. — Some abnormal sugar-canes. 

xxxix.-xlii. — Species of Stroplieodonta from Downing, N.S.W. 

xliii. — Ipsvicia jonesi and Osmylopsy chops spillerae. 

xliv. — Pherosphaera Fitzgeraldi. 

xlv.-xlvi. — Australian Myrmeleonidae. 

xlvii. — Au.stralian Mantispidae. 

xlviii.-l. — Secretory canals in Tristania conferta and Syncarpia laurifoUa. 



THE HIGH TEMPERATURE ORGANISM OF FERMENTING TAN-BARK. 

Part ii. 

By R. Greig-Smith^ D.Sc, Maeleay Bacteriologist to the Society. 

(Ten Text-figures.) 

[Read 28th March, 1923.] 

In the first part of this research, it was shown that tan-bark, after under- 
going a preliminary treatment, known as "tempering," was capable of being 
attacked by a thermophilic bacterium and during the fermentation a quantity 
of carbon dioxide was given off. It was suggested that the preliminary treatment 
eliminated inhibiting substances such as the residual tannins of the bark. 

On taking up the investigation again, the first problem that suggested itself 
was that of the nature of the fermentable constituents of the tempered bark. 
These are presumably soluble or slightly soluble in the liciuid in contact with the 
bark and it seemed advisable to make an aqueous extract and test its fermen- 
tability. 

Accordingly, 200 grams of tempered bark were warmed for 2 hours at 60° 
with 400 c.c. of water, and filtered through a Buchner funnel. The residue was 
again treated with water and this was repeated a third time. The total liquid 
and washings measured 1185 c.c. The fluid was evaporated under diminished 
pressure to 74 c.c. and filtered through a porcelain candle. 

Ten c.c. portions, approximately equal to the extract from 27 grams of bark, 
Avere put into flasks, one of which contained kieselguhr and another contained 
filter paper; a third served as a control. A fourth flask contained 30 grams of 
the dried residual bark. All four flasks were put into the apparatus described in 
the first part of this paper. It will be remembered that the flasks were kept at 
60° and a current of pure air swept any carbon dioxide evolved during fermen- 
tation into flasks of baryta water. 

The flasks were seeded with bacteria at the stai-t and were examined from 
time to time to determine the activity or vitality of the bacteria. Water was 
added as occasion demanded to replace that carried over in the current of air. 

An early question that arose was, how much carbon dioxide would one be 
justified in assuming as having been derived from a fermentation and how much 
from the errors of the apparatus? The error of the apparatus is traceable to 
leakage of air containing carbon dioxide into the apparatus, possibly through 
defective connections, possibly by diffusion through thin -walled rubber tubing 
and possibly to escape past the guarding soda-lime and baryta water. Against 



2 THE HIGH TEMPERATURE ORGANISM OF FERMENTING TAN-BARK, ii., 

this leakage there may have been a loss through the fermentation carbon dioxide 
having been imperfectly caught in the measured baryta water. Doubtless there 
was a certain amount of compensation one way and the other but some cases in 
which there could be no uncertainty about the absence of fermentation gave a 
yield of about one milligxam of carbon dioxide per diem. 

In the tables that follow, the daily determinations which were made to the 
first decimal place are omitted and only the aggregate product to the nearest 
whole number is given. This makes for conciseness and gives the data from 
which the curves were constructed. 

Table i. — Carbon dioxide in milligrams. 
Aggregate amounts. 

Days . 1 

1 . Extract with kieselgulir 1 

2. Extract with paper 5 

3 . Extract alone 2 

4. Water exhausted bark ^ . . . . 45 



2 


3 


4 


5 


6 


9 


— 


6 


16 


30 


40 


50 


— 


9 


13 


16 


— 


25 


— 


7 


l'( 


20 


— 


32 


135 


217 


282 


326 


372 


— 




DAYS 



The extract had undoubtedly been fermented, for the aggregate amount of 
carbon dioxide is well above the limit fixed for the error of the apparatus. But 
the fermentation had been feeble and it is evident that water extracted but little 
fermentable matter from the bark. The amount of extract in each flask was 
derived from 27 grams of bark and the 30 grams of the residual, water-exhausted 
bark gave from 10 to 20 times the amount of carbon dioxide in five days. 

The organic matter in the flasks was very small in amount. The liquid ex- 
tract, evaporated and dried at 130°, contained 2.5% of solids and 52.8% of this 
consisted of ash. Thus the 10 c.e. in the fermenting flasks contained 120 milli- 
grams of ash-free organic matter. 

The test-tube reactions for the brown-coloured extract were those of humic 



BY R. GREIG-SMITir. 3 

acid and doubtless this was the chief organic constituent. Naturally it would be 
present as humate. 

At the close of the experiment, the flask with the extract alone (i.e., No. 3) 
was kept and into it was placed a quantity of recently ignited willow charcoal. 
The idea was to see if the charcoal gave a better substratum for bacterial activity, 
something akin to moistened bark. 

Table ia. — CarDon dioxide in mgrms. 
Willow charcoal added to No. 3 on the ninth day. 

Days (continued) 10 11 13 15 17 19 23 26 30 33 36 

Aggregate yield starting from 32 ... . 47 55 63 73 80 85 94 105 117 124 130 



Before the addition of the charcoal, No. 3 ha.d given off: 32 milligrams. Next 
day 15 mgrms. were obtained, no doubt chiefly derived from carbon dioxide 
adsorbed by the charcoal and given off at 60°, From the jump of the curve it 
would appear that 10 mgrms. may be safely put down to this source and, making 
this allowance, it is shown that the charcoal aided the fermentation. We cannot 
go further and claim that the charcoal was itself fermented, for, although 130 
milligrams of carbon dioxide were obtained from 120 mgrms.* of humic acid plus 
charcoal in 36 days, yet the error of the apparatus, which we have put down 
at a milligram per diem, reduced the total to a quantity that might well be derived 
from the extract. It was, however, a trial experiment and as such is suggestive. 

In a second experiment, 15 grams of tempered bark passing through a No. 12 
sieve but retained by No. 24, were put into a Soxhlet oil-extraction apparatus 
and extracted with hot rectified spirit for 8 hours. On cooling, the alcohol threw 
out a fat which, on hydrolysis, jdelded stearic acid (m.p.70°). 

The alcoholic extract was evaporated, transferred to a flask with water, 
seeded with bacteria, mixed with ignited pumice fragments and connected up in 
the fermentation apparatus. 

The bark, after the alcohol treatment, was extracted with hot water in the 
Soxhlet apparatus for 14 hours. The aqueous extract thus obtained was 
evaporated and worked up as in the previous paragraph. 

The residual bark was percolated with 2% ammonia in the cold, then trans- 
ferred to the Soxhlet apparatus and extracted hot for about 12 hours. The ex- 
tract was evaporated to dryness, picked up with water and, being acid was 
neutralised with sodium hydrate and treated like the others. 

Table ii. — Carbon dioxide in mgrms. 

Days 1 2 4 5 6 7 9 10 11 13 14 17 18 19 20 21 23 24 

Alcoholic extract . . . . • — 2 — — — — 7 

Aqueous extract . . . . 10 13 — 22 — 31 — 47 58 — 63 — — 6S — 71 

Ammouiacal extract . . . . 6 17 29 — 42 — 59 — 68 80 — 92 — 108 115 — 122 — 

* This quantity of humic acid with 64 % of carbon is theoretically capable 
of yielding 281 mgrms. of carbon dioxide. 



THE HIGH TEMPERATURE ORGANISM OF FERMENTING TAN-BARK^ ii.- 




Tbe alcoholic extract gave onlj^ 7 mgrms. in 9 days and it is clearly evident 
that there was no fermentation. This is one of the examples that led to taking 
the error of the apparatus as one milligram per diem. 

The curve of the ammoniacal extract is almost a straight line^ from the second 
to the thirteenth day. The sudden jump on the seventeenth day may have been 
caused by the addition of 0.05 grams of peptone to the flask, although a similar 
addition to the aqueous extract on the eighteenth flay was without effect. 

The aqueous and ammoniacal extracts undoubtedly contained some material 
that was fermented. 

In determining the carbon dioxide caught in the baryta Avater by the difference 
in its alkalinity before and after exposure to the air coming from the flasks, it 
was necessary to allow for traces of ammonia, coming from some of the flasks, 
being caught in the cold baryta. At first this was done by neutralising the 
baryta, filtering off the barium carbonate, dissolving it in standard sulphuric 
acid, boiling to expel the carbon dioxide and determining the sulphuric acid 
neutralised. At a later date and especially when ammonium sulphate was used 
as a source of nitrogen in fermenting earbohj^drates, it was found to be more 
convenient to distil off from 10 to 20 e.c. into a flask containing 1 c.c. of O.IN 
sulphuric acid coloured with phenol-red. This gave an indication of the am- 
monia coming over and if the distillate remained yellow after 10 e.c. had passed 
over, tlie previously stopped inlet was connected with a soda-lime tube, the flask 
cooled down and the baryta titrated. When the distillate became alkaline, another 
e.c. of acid was added and the distillation was continued until it was certain that 
all the ammonia had passed over. 

A third experiment was made with the tempered bark. Twenty grams were 
digested for three days with 200 c.c. of 1% hydrochloric acid to remove calcium 
and magnesium salts. The insoluble matter was washed until free from chlorine 
and the filtrate and washings (A) were evaporated to dryness, moistened with 
water and again evaporated. This was repeated severaljimes and finally the 
.suspension was made neutral to litmus. It was transferred to a flask and tlie 



BY R. GREIG-SMITH. 5 

excess of water over 30 c.e. was removed by evaporation in a hot oven. The 
liquid was then seeded and the "flask put in the fermentation apparatus. The 
contents of the flask had to be neutralised daily for they rapidly became acid to 
dyes such as brom-thymol blue. Only a small amount of carbon dioxide was 
given off- — 15.5 mgrms. in 27 days — and this was taken to represent the error of 
the apparatus. 

The scheme of the experiment is tabulated in order to prevent confusion. 

20 grams of tempered bark treated with dilute hydrochloric acid, filtered. 

= i ~ ' ' ' ' 

Filtrate A i Residue treated -with dilute ammonia, filtered. 



Filtrate B i Residue again treated with ammonia, filtered.. 



Filtrate C 



Residue separated into 



Fine fragments D 



Coarse fragments E 



The residual bark after the acid treatment was digested in warm dilute am- 
monia (4%), allowed to stand overnight and filtered; it passed very slowly 
through the paper. Washing was continued until the filtrate became a pale 
brown. The filtrate and washings (B), after passage through porcelain, measured 
1614 CO. and this volume, less 10 e.c, contained 3.7 grams of dry matter, of which 
0.3 gram was ash. The fluid was evaporated, transferred to a small flask, seeded 
with the organism and put into the fermentation apparatus. 

In 1, 2, 4, 5, 6 and 7 days the aggTegate yields of carbon dioxide were 10^ 
17, 32, 37, 43 and 48 mgrms. 

As the fermentation appeared to be stopping on the 5th day, half a gram 
of peptone was added and, as this did not improve matters, the experiment 
was stopped on the 7th day. 

The contents of the flask were mixed with the acid extract (A) when the 
dark-coloured liquid was converted into a gelatinous paste, probably the humates 
of calcium and magnesium. In the fermentation apparatus a considerable amount 
of carbon dioxide was given off at first, but the production quickly fell away. 

In 1, 2, 4, 7, and 8 days the aggregate yields were 18, 21, 25, 36 and 38 
mgTms. of carbon dioxide. 

When tested on the seventh day, the pasty liguid was found to be sterile. In 
easting about for a reason for this state of affairs, it was remembered that the 
acid extract had been concentrated to a small volume and doubtless the salinity 
had become too great for the growth of the bacteria. The suspension was there- 
fore filtered and washed with a solution of 0.5% magnesium sulphate, a salt which 
had, by this time, been found to be very useful for the givwth of the bacillus. 
Thus the suspension was freed from any excess of saline matter aiid from peptone. 

The insoluble matter dried at 60° for two days weighed 4.7 grams. It was 
ground in a mortar and put into a flask together with 30 c.c. ammonium sul- 
phate solution* and seeded. 

* Ammonium sulphate 5, magnesium sulphate, anhyd. 1, sodium phosphate 2, 
potassium citrate 3 grams, brom-thymol blue solution 20 c.c. and tap-water to a 
litre. 



O' 



^ o 

L ! B R / 



THE HIGH TEMPERATURE ORGANISM OF FERMENTING TAN-BARK^ 11., 

Table iii. — Solid from Filtrates A -j- B . 
Carbon dioxide in mgrms. 



Days 

Aggregate yield 



2 3 5 7 9 11 13 16 18 24 28 

70 106 138 163 181 195 209 230 240 262 276 



200 




The liquid was neutral on the second day, but had to be neutralised on the 
eleventh and twenty-first days. The bacteria were active throughout the experi- 
ment. 

On plotting the yields, a curve is obtained and its logarithmic nature shows 
that the material had been fermented. A portion of the carbon dioxide would be 
derived from the citrate in the nutrient solution, but the yield is much in excess 
of the 73 milligrams theoretically capable of being derived from this source. 
Thus the insoluble matter obtained by adding the acid extract to the ammoniaeal 
extract and consisting presumably of the humates of calcium and magnesium, is 
fermented by the organism. 

The residue from Filtrate B Avas again treated with ammonia in much the 
same way as before and during the filtration, the insoluble particles were separated 
into heavier portions which remained in the beaker and lighter ones that were 
carried with the washing water on to the filter. 

The filtrate, C, when evaporated and dried, gave 1.8 grams of residual matter 
Avhich did not ferment. It became repeatedly acid during the attempted fermen- 
tation and, with the idea of preventing tliis, the acid Avas converted into a calcium 
salt by treatment with lime water. Even then there was no pronounced produc- 
tion of carbon dioxide — 39 mgrms. in nine days after allowing for included car- 
bonate — and, as citrate of potash was unfortunately present in the nutritive fluid, 
no deduction could be drawn regarding the fermentability of the humate. 



BY R. GREIG-SMITH. 



The smallei- fragments of the residual bark (D) weighed 2.7 grams. They 
were put into a flask together with 30 c.c. of ammonium sulphate nutrient solu- 
tion, Beutralised, seeded and fermented. 



Tatle iv. — iSmall fragments (D.), 
Carbon dioxide in mgrms. 



Days 1 

Aggregate yield 6 



3 


4 


6 


8 


10 


13 


15 


17 


25 


36 


61 


88 


116 


129 


136 


142 



As the citrate added with the nutrient solution could not give more than 73 
milligrams of carbon dioxide it is evident that the smaller fragments of residual 
bark were fermented. 

The eitrated nutrient solution had shown itself to be capable of producing a 
good growth of cells and certain control experiments had given no* carbon dioxide. 
Later experiments, however, showed that citrate could be and was fermented. 
The possibility of its being fermented in these experiments must therefore be 
taken into account. 

The coarse fragments of bark, E, after the second treatment with ammonia, 
weighed 6 grams and contained 14% of ash. A portion weighing 5.3 grams was 
put into a flask with 20 c.c. of water, seeded with the organism and adjusted 
in the fermentation apparatus. The fermentation began well but slowed down 
soon after the sixth day and half-a-gram of peptone was added in order to com- 
pensate for the loss of nitrogenous nutrients occasioned by the repeated extrac- 
tion of the bark with water. Then fermentation proceeded much more quickly 
but, as it was shown soon afterwards that peptone itself is fermentable, the in- 
creased yield of carbon dioxide may have been due to the added peptone. Ac- 
cordingly, on the nineteenth day, the solid contents of the flask were washed by 
decantation with water containing 0.5% of magnesium sulphate and the washed 
residue was put into a flask together with 20 c.c. of ammonium sulphate nutrient 
solution, seeded and connected up. Although the flask had no peptone, the 20 
c.c. of nutrient solution which was added contained 0.06 gram, of potassium 
citrate capable of yielding at most 49 milligrams of carbon dioxide and this 
should be allowed for in considering the fermentation. 

Table iva. — Coarse fragments of Bark (E) . Carbon dioxide in mgrms. Aggregate yield. 



Days 12 4 6 7 8 9 10 12 13 14 15 16 18 19 21 23 25 28 

With peptone . 18 27 48 54 62 81 102 121 148 159 — 178 — — 218 — — — — 
With citrate . . 17 — 31 — 51 56 — 69 83 — 97 — 116 130 — 152 168 183 205 

Considering the results from the bark with citrate and making an allowance 
for the carbon dioxide derivable from the citrate, it is evident that the coarse 
fragments of bark were fermented. 

In a fourth experiment, a thirty gram portion of ground and sifted alley- 
bark was digested with dilute hydrochloric acid at 80° for a short time and next 
morning was filtered and washed. The residue, after digestion with 4% am- 
monium hydrate for two days, Avas filtered through paper. A quantity of the 
filtrate was treated with a proportional part of the acid filtrate and evaporated 
to dryness, then picked up with water and filtered. The dried residue weighed 



8 THE HIGH TEMPERATURE ORGANISM OP FERMENTING TAN-BARK, U., 




DAYS 25 



3.5 gi-ams. It was placed in a distillation flask with 30 c.c. of a buffer mixture 
of inorganic salts and seeded after neutralisation with ammonia. 

Table v. — Humate in saline. Carbon dioxide in mgrms. 



Days 

Aggregate yield 



12 4 5 6 8 9 11 13 15 18 19 21 22 

3 12 22 28 33 43 49 57 61 70 88 94 112 123 




e (saline) 



20 DAYS 25 



BY R. GREIG-SMITH. 9 

The reason for making this test was that the humate in the earlier test 
(Table iii.) contained fermentable citrate, and the evolution of carbon dioxide, 
79 mgrms., by the second day seemed altogether too speedy for a true fermen- 
tation. It suggested an adsorption of carbon dioxide while the residue was dry- 
ing in the incubator. Assuming that the 79 mgrms. was adsorbed carbon dioxide 
and that the citrate yielded 70 mgrms. by the 16th day, the difference of 80 
mgrms. is not very far from the results of the present test. And as this test 
contained no fermentable substance other than the humate, and the fermentation 
started slowly and appeared normal, there is no reason to doubt the fermen- 
tability of humate. 

The speed of fermentation might have been faster had the buffer mixture of 
salts been less saline. It contained ammonium phosphate 0.5, potassium phos- 
phate 0.5, anhydrous magnesium sulphate 0.1, sodium chloride 0.1 and calcium 
chloride 0.1 gram in 100 c.c. of tap-water. The solution acted well when testing 
the reaction, giving the full depth of colour in a dilution of 1 to 70. But it 
was too saline for giving a maximum fermentation. In proof of this we have 
the following data of the fermentation of 0.1 gram of lactose. 



Days 1 4 5 

Buffer solution, 30 e.c 2 8 12 

Water, 15 c. c 

Buffer solution, 15 c.c 



} 



65 



The half-strength buffer solution gave a better fermentation than did 
Usehinsky solution containing ammonium sulphate in place of asparagin. Th^ 
latter gave an aggregate of 15 in four and 31 mgrms. in five days. 

After finding that the buffer solution of salts was too strong, the super 
natant fluid in the fermenting flask of the fourth experiment with humate was 
poured off and replaced by water. Th^ following amounts were obtained. 

Table va. — Humate in weak saline. Carbon dioxide in mgrms. 

Days 2 7 8 9 10 11 14 18 

Aggregate yield .... 7 29 44 53 62 90 135 189 

Up to the eighteenth day, the yield was at the rate of 10.5 mgrms. per diem 
as against 6 mgTms. in the denser fluid. 

In these experiments, the most pronounced fermentation was obtained from 
that humic acid which dissolved most easily in dilute ammonia. This did not 
occur, however, until the acid was combined with the bases of the bark previously 
extracted with dilute acid. The more difficultly soluble humic acid did not fer- 
ment to any extent but this may have been due to the absence of the natural 
bases. The organic matter insoluble in dilute ammonia was fermented, but not 
so actively as the humate. 

The complete elimination of humic acid from the bark is not easy. In the 
last experiment (see above) the residual bark was divided and a portion subse- 
quently found to be two-fifths (= 12 grams of original alley-bark) was treated 
with 4% ammonia. A deep-coloured extract was obtained. The residue was 
washed with dilute ammonia until the washings became of a port-wine colour. 
It was again treated overnight with 4% ammonia. This procedure was repeated 




10 THE HIGH TEMPERATURE ORGANISM OF FERMEXTING TAX-BARK, il., 

two times more. The amounts of humie acid dissolved were 0.25, 0.19 and 0.14 
grams by the third, fourth and fifth extractions respectively. It is evident that, 
for the complete elimination of. the humic acid, an eighth or ninth extraction 
would have been necessary. Thus in Tables iv. and iva., the fragments of ex- 
hausted bark would still contain humic acid and the carbon dioxide obtained by 
the fermentation of the fragments was probably derived from this residual humic 
acid. 

It will be remembered that after the bark has fermented in the stack, it is 
subjected to a process of "tempering"" during which it is watered and turned. 
The procedure palpably induces a mixed fermentation in which moulds play a 
part, for a white mouldy growth can be seen covering many of the outside lumps, 
from the results so far obtained, it is evident that "tempering" is a process of 
humdfication. 

There is certainly more humie acid in tempering allej'-bark than in tan- 
bark. As an example, a gram portion of dry tempered bark was treated with 
dilute hj^drochlorie acid and then with 50 c.c. of 4% ammonia for two days. The 
extract contained 0.46 gram of soluble matter. A control of dry tan-bark con- 
tained 0.32 gram. 

The flora of the tempering heap comprises bacteria, yeasts and moulds- and 
it is jDrobablj^ to the latter that the humification is mainly due. At any rate, 
among the moulds are several capable of fermenting cellulose and one would be 
justified in ascribing the degradation of the cellulose to these. 

A fresh specimen of tempering bark was used in making suspensions or 
dilutions of various strengths. These were smeared on plates of Czapek agar 
and gTown at 35°. On the second day the fourth plate contained 15 large 
colonies of which 7 consisted of a freely growing Actinomyces. 

There were three and a half millions of these in a gram of the moist bark. 
Other small colonies developed later, but the Actinomyces, by its rapid growth, 
was clearly an important constituent of the flora. It grew on eellulose-agar, dis- 
solving the cellulose and forming a clear halo around the margin of the growth 
or colony showing that it secreted an en'zyme capable of dissolving the cellulose 
in its vicinity. It seemed likely that this Actinomyces would play an important 
part in the decomposition or humification of the cellulose constituents of the 
bark. 

To test the matter, three two-gram portions of sterilised tan-bark were 
moistened and seeded, one with the Actinomyces, another with a suspension of 
alley-bark organisms, while a third served as a control. After a month's in- 
cubation at 35°, the barks were allowed to dry, and a fortnight later were trans- 
ferred to fermentation flasks, treated with 30 c.e. of ammonium phosphate saline, 
seeded with the high temperature organism and connected up in the thermostat at 
60°. At the time of determining the carbon dioxide, the contents of the flasks 
had generally a H-ion concentration of from 6.8 to 6.5 when a dilute solution of 
ammonia was added to bring them down to 7.2. 

Table vi. — Tan-bark seeded wiili Actinomyces and siiKponsion of .Mley-liarlr . Carbon 

dioxide in mgrms. 



Aggregate yields. 

Days 1 2 5 6 8 10 12 15 

Control 13 29 51 5ft 66 77 88 100 

Actinomyces sp. .. 12 26 48 54 63 72 81 93 
Allftv-bark suspension 15 31 56 65 82 94 100 125 



Approximate daily yields. 
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 
13 16 10 7 5 5 5- 5 5 5 5 5 4 4 4 
12 14 10 7 5 5 5 5 6 5 5 4 4 4 
15 IG 8 8 8 8 8 8 6 6 6 



BY R. GEEIG-SMITH. 



11 




ALLEY-BARK SUSPENSION 



ACTINOiViYCES & CONTROL 



Upon plotting the numbers of the aggregate yields, drawing the curves and 
deducing from these the daily yields, it is seen that Actinomyces had no influence 
in preparing the way for the high temperature organism. The test behaved 
very much as the control, indeed, after the second day the yields vvere practically 
the same. It seems to have rendered inert some of the more easily fermentable 
matter and a similar behaviour will be seen later in the fermentation of celMose 
(Table vii.). 

The bark that had been seeded with a suspension of organisms from alley- 
bark gave a greater amount of carbon dioxide than the control indicating that 
the mixed flora and fauna of tempered alley-bark contained an organism or mix- 
ture of organisms capable of altering the tan-bark so that a better fermentation 
was obtained. 

Other cellulose-dissolving moulds appeared upon plates of cellulose-ammonium 
suiphate-agar. They were vigorous growers and soon spread over the plates. 
One grew on potato, nutrient agar, salted paper, etc., as a white downy mass of 
interlacing aerial hyphae. These carried short septate conidiophores bearing a 
cluster of rather cylindrical spores measuring 2 x dfi- As a spore was formed 
at the pointed apex of the eonidiophore, it was pushed ofl: to make way for a 
younger cell and the spores were kept togetlier by a fluid. Thus each eonidio- 
phore was tenninated with a drop of liquid containing a number of spores. 
Microscopically, the aerial hyphae looked exactly like the block depicting 
Byalopus, the "Wasserkugelpilz," in P. Lindner's "Mikroskopische Betriebskon- 
trolle in den Garungsgewerben. 2te Auf., p. 233. 

On saline paper and all solid media, the mould grew as a white cotton-like 
vegetation, but here and there were small black specks. These proved to be 
aggregations of round black perithecia 100 — 120 ^ in diameter which, when burst 
by the pressure on the slide, released rounded asei (19/^) containing eight lenti- 
cular, dark-coloured ascospores (7.6 x 11.4;x). Among the asci were spherical 
cells (14/i) with granular contents. 

Another mould had a grey-green vegetation and grew easily on potato, agar 
of sorts, salted paper, tan-bark, etc. The aerial hyphae were arborescent m 
nature and each branch of the mycelial tree bore short septate conidiophores, 
terminated with a cluster of oval spores measuring 3 x 5/x. The clusters did 
not cai-ry a drop of water but the spores vrere kept together in little packets 
apparently by a film of some fluid. On plates of nutrient agar the mould grew 
as a fairy ring, vphich appearance was caused by the older conidiophores and 
aerial hyphae collapsing and tlie spores falling in a scattered heap on the agar 



12 



THE HIGH TEMPERATURE ORGAiSriSM OF FERMENTING TAN-BARK, li., 



floor. The mould slightly liquefied starch paste and produced some gas. It 
appeared to be a Gliocladium. 

A third was a grey-gTeen Aspergillus that grew freely on all ordinary media. 
The flask-shaped eonidiophores measured 22 to 25^, the basidia were 7.5^ long 
and the round smooth spores were 2.5^ in diameter. Like the other moulds it 
grew freely at 35° and did not liquefy starch j)aste. These characters, together 
with the simple nature of the sterigmata, indicated that it was Aspergillus fumi- 
gatus. 

These moulds during their groAvth dissolved the firiely divided cellulose as 
contained in eellulose-agar and, as the alteration of the cellulosic portions of tan- 
bark is the reason for the prelimmary treatment or tempering to which the bark 
is subjected, it was considered advisable to examine more closely their behaviour 
upon cellulose of a coarser texture. 

Accordingly use was made of filter paper. This was cut up into small 
squares and two gram portions were put into flasks together with 5 c.c. of am- 
monium j)hosphate saline* and after steaming were seeded with the moulds. 
After a month at 35°, the contents of the flasks were transferred to distillation 
flasks, 30 c.e. of the same saline added to each, seeded with the high temperature 
bacterium and maintained at 60°. 

Table vii. — Fermentation of Cellulose . Carbon dioxide in mgrms . 

Aggregate yields. Approximate daily yields. 



Days 1 

Control 2 

Actinomyces sp 3 

Hyalopus sp 13 

G-liocladium sp 25 

Asp. fumigatus 25 



2 


4 


6 


10 


20 


27 


5 


13 


20 


21 


30 


39 


3S 


47 


54 


35 


50 


54 



25 


10 


6 


5 


4 


4 


1 


2 


3 


4 


5 


6 


2 


9 


6 


4 


4 


3 


3 


4 


3 


3 


3 


3 


14 


7 


5 


5 


4 


4 


25 


11 


6 


5 


4 


3 




The approximate daily yields give a clearer indication of the progress of 
the fermentation than the aggregate yields. They show that Hyalopus, Glio- 
cladium and Aspergillus gave off a quantity of carbon dioxide during the first 
day when it was considered that the fermentation by the high temperature had 
scarcely had time to start. It would appear that the carbon dioxide was liberated 
by the moulds at a temperature of 60°, largely during the first day and slowly 



* Ammonium phosphate saline. Ammonium phosphate 0.3, potassium dihy- 
drogen phosphate 0.1, magnesium sulphate (anhydrous) 0.05, sodium chloride 
0.02.5, calcium chloride 0.02.5 grams in 100 c-c. of water containing 2 c.c of 
solution of brom-thymol blue. 



BY R. GEEIG-SMITH. 



13 



during the next three. The fermentation had its peak about the twelfth hour. 
In addition to this there is a fermentation of a constituent of the filter paper for 
the yields during the remaining days are greater than any error of the apparatus. 
The control indicates a fermentation with its peak on the second day, actually 
about the 36-hour period. From this the yield fell away very slowly in contrast 
with the rapid fall of the moulds. As the yields, in the Actinomyces test Avere 
less than the control, we must assume that some of the fermentable material was 
rendered inert or destroyed. 

Part of the experiment relating to the activity of Hyalopus, Gliocladium and 
Aspergillus was repeated. The moulds were gTown on the paper for thirteen 
d.a.js at 35° as against a month in the previous ease. 



Table viia.- 



-Fermentation of Cellulose. 
Aggregate yields. 



Carbon dioxide in mgrms . 

Approximate daily yields. 



Days 


1 


2 


5 


6 


7 


1 


2 


3 


4 


5 


6 


7 


Hyalopus sp. . . 


1 


2 


23 


27 


31 


1 


1 


9 


7 


o 


4 


3 


O-liocladinm sp. 


6 


20 


33 


36 


40 


6 


13 


6 


4 


4 


3 


a 


Asp. fumigatus 


. 13 


19 


32 


35 


38 


12 


7 


5 


4 


3 


3 


s 




3 4 5 6 DAYS 7 

There are some differences in these results and it is possible that the shorter 
incubation of the moulds is responsible. Aspergillus had its peak on the first 
day, Gliocladium on the second and Ht/alopus on the third day. As each received 
the same amoimt of bacterial suspension it would seem that the shorter incubation 
gave products of variable fermentability. 

The action of the high temperature organism upon the paper attacked hj 
the moulds called for some elucidation, especially in the rapid evolution of carbon 
dioxide during the first few days and the long continued but slow later fermen- 
tation. It seemed possible that the keeping of the paper in the moist sterile 
condition might either induce the absorption of some atmospheric carbon dioxide 
or might so alter the paper that it could be easily attacked by the high tempera- 
ture organism. The mould in such a ease might have little influence in prepar- 
ing the paper for attack by the bacillus. 

Some exi^eriments were made to determine these points. In the first, water 
and paper and also ammonium phosphate-saline and paper were kept for 24 daj^s 
at 35° when, they were transferred to the fermentation apparatus. 

The water and paper gave 3 mgrms. of carbon dioxide in three days and 
no more by the sixth day; water alone gave 2 mgrms. by the third and no more. 
The saline and paper gave 4.5 mgi-ms. by the ninth day and the saline alone 
gave the same. Thus about half a milligram was given off daily and this pro- 
TDably represents the error of the apparatus. "We may therefore conclude that 
-there is no actual fixation or absorption of carbon dioxide by paper or by 
nutrient fluids during nearly a month's storage in the incubator at 35°. 

On the ninth day the saline and the saline plus paper Avere seeded with the 
high temperature organism. 



14 THE HIGH TEMPERATURE ORGAKISM OF FERMENTING TAN-Bi^RK, ii., 

Table viii. — The Fermentation of Cellulose by the Bacillus. Carbon dioxide in mgrms. 



Days 



Aggregate yields. 
Sterile. Seeded. 

3 6 9 10 11 13 14 15 17 20 



Approximate daily yields. 
Seeded. 
10 11 12 lb 14 15 16 17 



Saline and paper 2 3 5 
Saline alone . . 2 3 5 



6 14 21 24 26 30 35 

6 • — 7 — — 



.0 4.5 3.5 3.3 2.4 



.0 1.5 



Following the addition of the high temperature organism to the paper a 
quantity of. carbon dioxide was given off; the yield rose to a maximum on the 
11th day, two days after the seeding. It then fell away slowly until the 17th 
day when the yield was 1.5 mgrms. at which it remained until the 20th day when 
the experiment was stopped. 

This experiment is instructive in showing that paper can be fermented by 
the high temperature organism. The control jdelded 7 mgrms. in 15 days aiid 
the paper yielded 26 mgrms.; the difference of 19 mgrms. of carbon dioxide re* 
suited from the fermentation of the paper. It was not very much, but still it 
was a fermentation and it showed that the carbon dioxide produced from the 
fermentation of paper after "tempering" by moulds was in part due to the fer- 
mentation of the cellulose directly. 

In order to see if carbon dioxide was given off when the gTowth of mould 
upon paper was subjected to a temperature of 60° in a current of air for some- 
days, an experiment similar to vii. and viia. was started. After gi-owing at 35° 
for 16 days, the cultures along with the paper and fluid were transferred to dis- 
tillation (fermentation) flasks and maintained at 60°. When tested on the third 
day for the high temperature bacillus, the fluids in the flasks containing Asper- 
gillm and Hyalopiis were found to be infected. The experiment, therefore, so 
far as they were concerned, became a duplication of vii. and viia. Two odd 
tests of Aspergillus and Hyalopus which were also found to be infected are in- 
cluded in the table. 

Table viib. — Fermentation of Cellulose after attack by Moulds. Carbon dioxide in mgrms. 

Days X 

Hyalopus sp,, 16 days 1 

Hyalopus, sp., 14 days 7 

Asp. fumigatus, 16 days 5 

Asp. fumigatus, 9 days 2 

Controls without paper — 



2 


3 


4 


3 


14 


24 





39 


50 


18 


25 


30 



A culture of Gliocladium on two grams of filter paper in 30 c.e. of am- 
monium phosphate saline, after 10 days at 35° was transferred to a fermen- 
tation fla,sk and maintained at 60°. When transferred, the fluid in contact with 
the paper had an acidity of pH 4.4 and both it and the control were brought to 
pH 7.0. On the third day when the calculated daily yield had fallen to zero and 
the flask had been found to be free from the high temperature bacillus, a culture 
of the organism was added to each flask. 

Table ix. — The Fermentation of Cellulose after attack by Gliocladium. C:ir1)on dioxide 

in mgrms. 



Days 

Gliocladium, saline and paper 
Gliocladinm and saline . . . 



Aggregate yields. 

Seeded with bacillus. 
13 4 5 7 8 9 11 14 
4 6 66 89 107 113 117 123 134 

0.3 2 4 — — — 6 — — 



Approximate daily yields. 

3 4 5 6 7 8 9 10 11 
60 23 11 7 6 4 3 3 



BY R, GREIG-SMITH. 



15 




In the fermentation of the paper previously attacked or "tempered"" by 
Gliocladium, we have an example of a comparatively rapid fermentation which 
shows that the substances formed by the mould from the paper are of an easily 
fermentable nature. This was exhibited by the jump from zero to 60 mgTms. 
during the 24 hours following- the introduction of the bacillus. The important 
point of the experiment is that the mould attacks cellulose and forms substances 
which do not liberate carbon dioxide on warming and which are attacked by the 
high temperature organism. The nature of these substances is at present un- 
known and although the liquid in contact with the mould became acid, tliere 
was not enough organic acid formed, as was shown by later tests, to ascribe the 
fermentation to the mould. Moulds are known to attack cellulose producing 
eellobiose and it vvas doubtless this sugar that v*as fermented. 

A further experiment on the same lines was made, the moulds being grown 
in contact with the paper for nine days at 35°. After two days in the fermen- 
tation apparatus at 60° they were tested for the presence of the high temperature 
organism and then seeded with it. The Asp. fumigatus test was found to be 
contaminated and has been included in Table viib. The other mould cultures 
were sterile so far as the bacillus was concerned. 

Table ixa. — The Fermentation of Cellulose after attack by Moulds. 
Carbon dioxide in mgrms. 



Aggregate yields. 

i Seeded with bacillus. 

3 4 5_ 7 

8 13 15 19 

Hyalopus sp a 2 | 22 31 39 47 

Gliocladium sp 3 3 I 10 14 18 21 



Approximate daily 


yields 


Seeded. 




J. 2 3 4 .=5 


fi 7 


10 7 4 3 


2 2 


2 8 5 3 


•2 1 


3 19 11 7 


4 3 



These bear out and confirm the previous test made with Gliocladium show- 
ing that the moulds attack the cellulose and the products are fermentable by the 
high temperature organism. During the attack by the moulds the paper became 



16 THE HIGH TEMPERATURE ORGANISM OP FERMENTING TAN-BARK, ii. 

pulpy and, microseopieally, many single fibres were seen to have been split up 
^nto numerous strands. 

The fermentable substances appear to be dissolved in the fluid in contact 
with the paper, for 30 c.c- of the fluid of a seven day's bulk culture, when seeded 
with the bacillus, gave 13, 7 and 3 mgrms. of carbon dioxide in 1, 2 and 4 days 
respectively. It behaved like the cultures in the Tables ix. and ixa. 

In the tempering heap where a revivification of the spent bark takes place, 
it is evident to the eye that moulds probably take an active part in the process. 
The characteristic fermentable substance of tempered bark is humie acid and ex- 
tJ-^aets containing this were shown to be attacked by the high temperature or- 
ganism. There is more humate in tempered bark than in tan-bark and the tem- 
pering may be said to be a humifieation. 

Several moulds were isolated from the bark and these attacked filter paper, 
giving rise to substances which were fermented by the thermophilic bacillus to 
carbon dioxide. 

My thanks are due to Mr. W. W. L'Estrange for much kindly assistance 
in the work. 



17 



STUDIES IN AUSTRALIAN ENTOMOLOGY. 

No. xviii. New Genera and Species of Carabidae. 

(Scaritini, Pterostichini, Merizodini, Bembidiini, Trechini, Odacanathini, 
Panagaeini, Licinini, and Lebiini.) 

By Thomas G. Sloane. 

{Continued from Vol. xl., 1915, p. 473*) 

[Read 28th March, 1923.] 

Tribe iSciaritini. 

Genus Dyschirius. 

Table of Australian Species. 



1 


(2) 


2 


(1) 


3 


(4) 


4 


(3) 


5 


(8) 


6 


(7) 


7 


(6) 


S 


(5) 



mastersi Mael. 
stephensi Mael. 



Colour uniformly black. Length, 2.5 mm 

Colour not wholly black. 

Elytra reddish testaceous with a broad central black fascia. Length, 

3.5 mm , . . . zonatus. 

Elytra never with a central black fascia. 

Size major. Length, 4 mm. 

Head and pronotum (except peduncle) ferruginous red 

torrensensis Blackb. 

Head (except clypeus) and pronotum black macleayi Sl. 

Size minor. 
9 (10) Black; elytra with two sanguineous maculae behind middle; striae not 

extending behind transverse basal depression. Length, 2-2.3 mm 

ovensensis Blackb. 
10 (9) Prothorax piceous black, elytra ferruginous brown; striae extending back- 
wards to ante-apical maculae. Length, 2.5-3 mm wilsoni SI. 

In arranging this table I have not had before me D. stephensi Mael., D. 
zonatus Putz., and D. torrensensis Blackb., but have relied on the descriptions of 
these species for the differences mentioned. 

Dyschirius wilsoni^ n.sp. 

Elongate-oval, convex. Piceous, head and pronotum black; elytra ferrugin- 
ous brown with a transverse reddish macula on each eljrfcron above apical de- 
clivity; under surface of a dull reddish colour; antennae infuseate with basal 
joint livid; legs reddish, apex of femora, base of tibiae, and tarsi infuseate. 

Vertex smooth, convex; front defined posteriorly by a transverse straight 
ridge turned obliquely backward near each eye, a fine longitudinal median ridge, 
a transverse sulcus (on each side of median ridge before the posterior ridge), 
and a juxta-ocular sulcus on each side; clypeus declivous, emarginate, lateral 

*There is no number xiii. in this series, so that No. xiv. follows No. xii. 



18 STUDIES IN AUSTRALIAN ENTOMOLOGY^ xviii., 

angles slightly prominent, supra-antennal plates large, convex. Prothorax glo- 
bose, pedunculate, laevigate, nitid; basal channel defining peduncle. Elytra trun- 
cate-oval, very convex, lightly transversely impressed at basal third, punctate- 
striate on basal half (before maculae); striae shorter on sides than on disc; 
interstices a little convex before transverse impressions, 3, 5, and 7 seriately 
setigero-punctate ; lateral channel narrow, punctate, wider and with larger punc- 
tures near base; border narrow, forming a minute dentieule at humeral angle. 
Anterior tibiae with apical digitation long, narrow, pointed, two outer teeth 
small, dentiform. Transverse suture of metasternum marked by a row of six 
large punctures. Length, 2.5 — 3 mm; breadth, 0.9 — 1.05 mm. 

Hah. — Victoria: Beaconsfield, near Melbourne. Several specimens were 
taken by Mr. F. E. Wilson in tussocks of grass — Colls. Wilson and Sloane. 

Closely allied to B. stephensi Macl., and D. ovensensis Blackb., which it re- 
sembles in form of head and prothorax. From D. stephensi and D. mastersi 
Macl. it differs by colour not wholly black; the prothorax is proportionally widei 
and more rounded on sides than in D. stephensi. From D. ovensensis it differs by 
colour; elytra with striae reaching back to the maculae. In D. ovensensis the 
inner striae are shorter than the outer ones. It is at once differentiated from D. 
torrensensis Blackb., and D. macleayi SI. by its smaller size. 

Genus C l i v i n A. 
Clivina sulcaticeps, n.sp. 

Narrow, cylindrical. Head narrow (1.3 mm. across eyes), depressed, longi- 
tudinally trisulcate on each side of vertex, clypeus widely and decidedly emar- 
ginate, not divided from the rounded lateral parts, mandibles long, prominent, 
decussate, labrum 5-setose; prothorax longer than broad, evidently narrowed to 
apex; elytra parallel, striate, stria 4 joining 5 at base, interstice 8 earinate at 
base, not raised at apex; peduncle with lateral cavities punctate; prosternum with 
intercoxal part wide, pro-episterna transversely rugulose and punctate; anterior 
tibiae 4-dentate, lower side with a strong transverse ridge beneath second ex- 
ternal tooth. Castaneous, head and prothorax darker than elytra. 

Head elongate, impunctate, nitid: clypeus smooth, strongly divided from 
front in middle, raised into a transverse ridge between frontal impressions; 
lateral parts depressed, not defined from median part or supra-antennal plates, 
rounded and decidedly more prominent than median part : eyes small, round, 
lightly convex. Prothorax narrow (2 x 1.7 mm.), widest near posterior angles, 
lightly narrowed to apex (1.5 mm.), convex; disc laevigate, subrugulose on sides; 
anterior margin lightly emarginate in middle; anterior border strongly developed; 
anterior angles near head, narrow, prominent; posterior angles rounded; median 
and anterior lines strongly impressed. Elytra cylindrical (4.5 x 2 mm.), deeply 
striate; striae crenulate. Length, 8.8; breadth, 2 mm. 

Hah. — Northern Territory: Alligator River. Type in National Museum, 
Melbourne. 

It has seemed better not to formulate a new genus for this aberrant species, 
but to leave it in Clivina, where it will form the type of a new group. The form 
of the anterior tibiae, with their transverse ridge on lower side, has not been seen 
by me elsewhere in the genus Clivina, but occurs in the South American genus 



BY T. G. SLOANE. 19 

Oxystomus, while the striated head recalls that of Schizo genius. The form of the 
maxillary palps and mentum resembles those of Clivina planiceps Putz., but the 
mentum has not the longitudinal ridges of the lobes as in that species. 

Genus Laccoscaphus. 
' Laccoscaphus ctaneus Fabr., var. AE^^ESCENS; n.var. 

Differs from the typical form of L. cyaneus Fabr., by having the upper sur- 
face of a coppery bronze colour. Length, 10,5 — 12 mm. 

Hah. — Queensland: Herberton District (Dodd). Colls. Sloane (type) and 
British Museum (cotype). 

Note. — L. cyanevs varies considerably in size; specimens in my possession 
measure from 10.5 to 14 mm. in length. It ranges from Cairns at least as far 
north as Cooktown, and extends inland at least to Chillagoe. 

Genus Teratidium. 
Teratidium parallelum, n.sp. 

Elongate, cylindrical. Head convex, subquadrate, constricted posteriorly; 
posterior part of frontal impressions obsolescent. Mtid, black, elytra vireseent 
with purple reflections; prothorax with slight viridescent and purple reflections 
towards base aad sides. 

Head large (4.5 mm. across eyes), two supra-orbital punctures on each side; 
front with a foveiform impression at each end of clypeal suture; occiput de- 
pressed below plane of vertex; clypeus strongly declivous, median part roundly 
emarginate; labrum emarginate; eyes deeply set in and not more prominent than 
orbits; orbits large, projecting strongly and almost rectangularly from sides of 
head, almost parallel on outer side, enclosing posterior two-thirds of eyes. Pro- 
thorax longer than broad (5.2 x 4.5 mm.), lightly angustate to base; sides 
parallel, lightly angustate just before base ; anterior margin truncate, angles 
lightly and obtusely advanced; base arcuate, angles obtuse; border narrow, 
thicker and more raised on base; marginal channel narrow, bearing 2 or 3 punc- 
tures on anterior half and 2 punctures at beginning of basal curve on each side. 
Elytra laevigate, a shade wider than prothorax (10 x 4.6 mm.), a little depressed 
on disc, parallel; base emarginate, steeply declivous; lateral border thick and 
convex posteriorly, narrow upturned and rounded at humeral angles; a wide 
oT)lique punctate post-basal impression on each side. Yentral segments unipune- 
tate on each side. Legs light; anterior tibiae with apex prominent, slender; a 
smaU dentiform prominence on outer side at base of apical digitation; posterior 
coxae and trochanters impunctafe. Length, 20; breadth, 4.6 mm. 

Hob. — Northern Territory: Adelaide River. Type in British Museum. 

It belongs to th« section of the genus in which the front is without strong 
sulci between the eyes. It is nearly allied to T. eonvexum SI., the chief differences 
being the viridescent colour of the elytra, clypeus semicircularly emarginate be- 
hind the labrum (which is also roundly emarginate), orbits standing out more 
abruptly from head; prothorax with anterior angles more obtuse; elj'tra more 
parallel and more obtuse at apex; anterior tibiae wider, and with the small ex- 
ternal dentieule more developed. 



20 STUDIES IN AUSTRALIAN ENTOMOLOGY^ xviii., 

Tribe PteFOfstichini. 

Phersita crenulata_, n.sp. 

2. Elliptical, rather depressed. Prothorax subquadrate, wider at base 
(1.85 mm.) than apex (1.4 mm.) ; elytra ovate, crenulate-striate, interstice 1 with 
a short striole at base, 3 unipunctate beside stria 3 at middle, 8 carinate towards 
apex, basal border dentate at shoulders. Piceous red. 

Head convex, rather narrow (1.3 across eyes); front lightly and shortly bi- 
impressed, median space lightly convex; eyes (including orbits) reniform, small, 
convex but not prominent. Prothorax broader than long (1.7 x 2.1 mm.), 
widest before middle; sides lightly arcuate on anterior half, very lightly narrowed 
posteriorly, straight before base; apex lightly emarginate, angles obtuse, bordered; 
base wide, truncate, bordered, angles rectangular; basal area depressed, rather 
densely but finely punctate, the puncturation stronger in basal impressions; these 
wide, shallow, flat; lateral border narrow, reflexed; a narrow subeostate space 
dividing basal impressions from marginal channel. Elytra truncate-oval (4 x 2.8 
mm.), lightly and widely convex; interstices depressed on disc, 7 convex; stria 7 
fully developed. Ventral segments 4 — 6 with a row of closely set fine punctures 
along anterior margin, 6 plurisetose at apex. Length, 7.2; breadth, 2.8 mm. 

Hab. — N.S. Wales: Mount Kosciusko. I found a single specimen near the 
Hotel Kosciusko (5,000 feet) on 6th December. 

A distinct species. From Ph. melbournensis Cast., it differs by head nar- 
lower, eyes not nearly so prominent; prothorax longer, less rounded on arcuate 
part of sides, more lightly narrowed to base in a longer and more gentle slope, 
basal area more depressed and more punctate, basal impressions shallower and 
flatter; elytra proportionately Jlonger, less convex, striae finer and shallower on 
disc, etc. From Ph. tasmanica SL, it can be distinguished, inter alia, by prothorax 
less rounded on sides, basal area larger, more depressed and punctate. 

Genus Pbosopogmus. 

Prosopogmus opacidermis, n.sp. 

Elliptical, subdepressed. Prothorax angustate and sinuate posteriorly, wider 
at base (2 mm.) than apex (1.7 mm.); basal angles rectangular with apex 
blunted; elytra fully but not deeply striate, interstices (including 8) depressed, 
5 — 7 narrower and lightly convex at apex, 3 tripunctate, 8 free at apex; met- 
episterna longer than broad. Black; head and prothorax nitid; elj^ra subnitid in 
6, subopaque in ?; legs black, tibiae piceous; antennae, palpi and tarsi piceous 
red. 

Head not large (1.65 mm. across eyes), convex; frontal impressions well 
mai-ked; eyes reniform, rather prominent. Prothorax broader than long (c?, 
2x2.3 mm.), lightly sinuately narrowed to base; sides lightly arcuate anteriorly; 
anterior angles obtuse, subprominent, bordered; base truncate; inner basal im- 
jjression striolate, rather long, outer impression striolate, feeble, interspace flat; 
lateral border narrow, ending at basal angle; a seta beside basal angle inside 
border. Elytra oval (5x3 mm.) ; basal border a little raised above lateral 
border and feebly dentate at humeral angle; anterior puncture of interstice 3 be- 
side stria 3, two posterior punctures beside stria 2; interstices 5 and 7 enclosing 
6 at apex, 7 — 9 nearly equal in width. Prostemum bordered at base of inter- 
coxal part. 



BY T. G. SLOANE. 21 

?. More robust than c?; pro thorax wider (2 x 2.5 mm.); elytra duller in 
colour. 

Length, 8.5 — 9.5; breadth, 3 — 3.5 mm. 

Haft. — N.S. Wales: Eecleston (J. Hopson). 

This species can only be confused with P. monochrous Chaud., which it closely 
resembles in size and appearance, but it differs by prothorax narrower at base 
and more sinuate on sides; elytra less nitid in both sexes (especially ?), ii;ter- 
stices less convex in both sexes, depressed on disc in ?, shagreened in both sexes 
(so strongly so as to be opaque in 2), third interstice 3-punctate. 

Ncte. — P. opacidermis is the only species of the genus in which I have been 
any variation in the number of punctures on the third interstice of the elytra; 
eight specimens are before me; of these, six have the third interstice normally 3- 
punctate, as described above; one of the remaining two specimens (?) has the 
left elytron normally punctate, but only the middle puncture present on the right; 
the other has the two anterior punctures present on the left el.ytron, but only the 
middle puncture on the right. 

PROSOPOGMUS INTERSTITIALIS, n.Sp. 

Parallel, elliptical. Prothorax subquadrate, wider at base (2 mm.) than 
apex (1.6 mm.), strongly bi-impressed on each side of base, basal angles almost 
rectangular; elytra strongly striate, _ interstices convex, 3 tri-punctate, 8 free at 
apex, met-episterna elongate. Black, nitid; femora pieeous; tibiae, tarsi, and an- 
tennae piceous brown. 

Head moderate (1.5 mm. across eyes), convex; frontal impressions well 
marked, short, extending across clypeus, divergent backward; eyes prominent, 
lightly enclosed behind; post-ocular parts of orbits small, but projecting sharply 
from head. Prothorax broader than long (1.8 x 2.2 mm.), widest a little before 
middle, depressed on disc; sides lightly curved, oblique near base; anterior angles 
narrow, obtuse, rather prominent; base lightly emarginate in middle, angles sub- 
rectangular with apex blunted; inner basal impressions subfoveiform, outer im- 
pressions f oveate, deep ; sometimes a few punctures in basal impressions ; a seta 
just within basal angle. Elytra truncate-oval (4.2 x 2.75 mm.); hujneral angles 
strongly marked, dentate; interstices convex, 1 — 4 more convex in d* than ?, 3 
with anterior puncture beside stria 3 and the two posterior ones beside stria 2, 
6 — 8 enclosing 7 at apex, 7 — 9 subequal in width; a short striole at base of inter- 
stice 2. Prostemum bordered at base. Length, 7.2—7.5; breadth, 2.75 mm. 

Hah. — N.S. Wales: Mount Wilson (Carter); Eecleston. 

Closely allied to P. ooddformis Macl., from which it differs by elytra nitid 
(not shagreened), black (not olivaceous) ; prothorax proportionately narrower, 
especially towards base, basal impressions deeper; elytra more deeply striate, 
interstices convex, etc. Formerly I confused this species with P. occidentalis 
Macl., but that species is smaller, and has the basal foveae of the prothorax more 
punctate. In some lights a slight sub-violaceous flush may be noticed on the 
elytra, especially in c?. I believe P. inter stitialis extends from the Blue Moun- 
tains into Queensland, perhaps as far as the tropics. 



22 STUDIES IN AUSTRALIAN ENTOMOLOGY^ Xviii,, 

Genus T ri oho sternu s. 
Teichosternus perateb^ n.sp. 

Elongate-oval. Head bisetose on each side abave eye, tooth of mentum 
rounded, penultimate joint of labial palpi with two median setae and a setule at 
inner side of apes; prothorax transversely subcordate, lateral margins bisetose, 
posterior seta a little before basal angle; elytra oval, costate, humeral angles pro- 
minent; j)rosternum plurisetose between coxae, intercoxal declivity concave, njeso- 
sternum glabrous between coxae. Black. 

Head large (8.5 across eyes), swollen below eyes; genae arcuate-truncate 
outside base of mandibles, outer angle obtuse but marked; frontal impressions 
wide, shallow. Prothorax broader than long (7.75 x 10.3 mm.), nitid; surface 
with faint transverse striolae; sides arcuate on anterior two-thirds, lightly sinuate 
posteriorly and meeting base at right angles; apex truncate, angles obtuse, not 
prominent; base emarginate in middle, lightly arcuate on each side; basal angles 
rectangular, sub-prominent and obtuse at summit; lateral border thick, wide near 
anterior angles; median and anterior transverse lines lightly impressed; lateral 
basal impressions wide, foveiform, connected by a well-defined transverse im- 
pression. Elytra oval (26 x 13 mm.), lightly convex, lightly and evenly rounded 
at sides, subopaque, finely shagreened ; striae feebly punctate ; odd interstices 
strongly raised (7 strongest), nitid at summits, even ones subconvex, opaque, 3 
quadri-punctate on posterior two-thirds, 8 not indicated at apex, 9 a little raised, 
seriate-punctate; basal border decidedly but obtusely raised at humeral angles. 
Metasternum not punctate at sides. Abdomen in c? 2-, in ? 4-setose at apex. 
Posterior coxae contigTious. <S, anterior tarsi lightly dilated, weakly squamosa 
beneath thi-ee basal joints. Length, 36 — 38.5; breadth, 12 — 13 mm. 

Hah. — N.S. Wales: Tweed River (Carter and Sloane). Coll. Sloane. 

This fine species was first given to me by Mr. H. J. Carter who had taken 
it on the Tweed River; afterwards I got a single specimen in the brush near 
Murwillumbah. In size it equals T. renardi Chaud., which it resembles so closely 
that at a casual glance it seems to be conspeeific, but an examination shows that 
it differs greatly by intercoxal' declivity of mesosternum glabrous; external angles 
of genae far more marked (in T. renardi the outer side of the genae is similar in 
its curvature to the lobes of the mentum, in T. perater the outer side is squarer 
and more truncate at apex) ; prothorax more ampliate anteriorly, anterior angles 
more widely margined and therefore more distant from neck, transverse im- 
pression connecting lateral basal impressions deeper; elytra with eighth inter- 
stice not developed at apex, ninth less clearly defined from lateral channel, punc- 
ture at base of first interstice wanting; prosternum not longitudinally channelled, 
but with a wide foveiform impression between coxae; metasternum not punctate 
at sides; posterior femora narrower, less roundly inflated on lower side; c?. three 
basal joints of anterior tarsi far less strongly dilatate, and less squamose beneath. 

Trichosternus simplicipes, n.sp. 

Elongate-oval, subd^ressed. Head lightly bi-impressed, bisetose on each 
side above eyes, tooth of mentum rounded at apex, penultimate joint of labial 
palpi bisetose; prothorax truncate-cordate, lateral margins bisetose, posterior seta 
near ba«al angle; elytra oval, interstices 1—7 costate, 3, 5 and 7 more raised 
than others, humeral angles strongly and obtusely dentate; intercoxal parts of 



. . BT T. G. SLOANE. 23 

prosternum and mesosternum setose; 6. anterior tarsi naked beneath. Black, pro- 
thorax virideseent on sides, elytra with ninth interstice and lateral channel 
cupreous. 

Head large (5 mm. across eyes); frontal impressions wide, shallow; orbits 
shortly swollen behind eyes; genae projecting widely from base of mandibles, 
subparallel at sides, arcuate at apex. Prothorax broader than long (4.3 x 6.3 
mm.), widest just behind anterior marginal seta; sides lightly rounded anteriorly, 
lightly sinuate posteriorly, meeting base at right angles; apex lightly emarginate, 
angles obtusely rounded, hardly advanced; base emarginate and fringed above 
peduncle, truncate on each side, not bordered , angles subrectangailar ; lateral border 
reflexed; lateral channel narrow; median line lightly impressed; lateral basal 
impressions elongate, shallow; spaces between these impressions and border flat. 
Elytra oval (12 x 8 mm.), truncate at base, lightly declivous to sides and apex; 
striae punctulate; interstices 3 and 5 strongly costate, 7 carinate, 2, 4 and 6 
lightly costate, 3 bi-setose on apical half, 9 nitid, narrow, punctate, developed 
towards apex. Metasternum with one or two punctures near each posterior 
angle. Abdomen in <S 2-, in ? 4-setose at apex. Posterior coxae contiguous. 
Length, 22—24; breadth, 7.4—8.5 mm. 

Hob. — Queensland: Bunya Mountain (H. J. Carter). Four specimens have 
been examined; given to me by Mr. H. J. Carter, who found this species not un- 
common at the Bunya Mountain, South Queensland, in October. 

Allied to T. suhvirens Chaud., with which it agTees in the form of tooth of 
mentum, and in secondary sexual characters; but differing by form narrower; 
head black; prothorax less metallic; elytra black (excepting margin, cupreous), 
interstices of disc more unequal, the even ones being less costate than the odd 
ones. 

Trichosterxus setosiceps, n.sp. 

+. Elongate-oval. Head strongly bi-impressed, 4-setose on each side above 
eye, tooth of mentum excised at apex, penultimate joint of labial palpi 3-setose 
on inner side and with two distal setules; prothorax subeordate, disc opaque and 
transversely striolate, lateral margins bisetose, posterior seta considerably before 
base; elytra 3-carinate, humeral angles hardly prominent; intercoxal part of 
prosternum 6-setose, of mesosternum glabrous. Black. 

Head large (5.3 mm. across eyes); sides lightly tumid behind eyes; genae 
projecting widely from base of mandibles, subparallel at sides, roundly truncate 
at apex; prothorax broader than long (4.5 x 6.4 mm.), widest at anterior mar- 
ginal seta; sides rounded on anterior two-thirds, sinuate posteriorly and meeting 
base at right angles; apex lightly emarginate; anterior angles obtuse, a little ad- 
vanced; base narrowly bordered, lightly truncate-emarginate and with a hair- 
fringe above peduncle, truncate on each side; basal angles rectangnilar, obtuse at 
summit; lateral border nitid, its course interrupted by anterior setiferous punc- 
ture; lateral channel wide, finely granulate, opaque; median and anterior trans- 
verse impressions strongly marked; lateral basal impressions strongly impressed. 
Elytra opaque, oval (12 x 8 mm.), a little narrowed to base, lightly convex, 
strongly declivous at sides; striae distinct, punctate; interstices 3, 5 and 7 
carinate (summits nitid), 2, 4, 6 and 8 depressed but a little convex posteriorly, 
9 merged with margin; lateral margin wide, nitid; basal border strong, con- 
necting bases of interstice 7 and latera.l border with a little nodule at hmneral 
angle. Posterior coxae contiguous. Abdomen 8-setose at apex. Length, 20.5 — 
23; breadth, 6.6 — 8 mm. 



24 ■ STUDIES IN" AUSTRALIAN" ENTOMOLOGY^ Xviii., 

Hah. — Queensland: South Johnstone River (H. W. Brown); Malanda (G. 
F. Hill). Two specimens (?) are before me, one is in my collection, the other 
belong-s to Mr. G. F. Hill. 

It resembles T. marginatus Chaud., in appearance, but is quite distinct from 
it, and from all other described species, by head with four supra-orbital setae on 
each side, prothorax opaque and transversely striolate as in Loxogenius opaci- 
pennis Macl. Its position in the genus is beside T. cyaneocinctus Boisd., and T, 
superhus Cast. 

NOTONOMUS DEHISCENS, n.Sp. 

Oval, depressed. Head stout; prothorax cordate-quadrate, sinuate on sides, 
basal angles subrectangular, posterior marginal seta in lateral channel at basal 
angle; elytra complanate, fully striate, interstices lightly convex in J', depressed 
in ?, 3 with three or four punctures, humeral angles rounded. Black (rarely "with 
a "vireseent flush on prothorax and elytra); antennae piceous; legs, basal joint of 
antennae, and mouth parts piceous red. 

Head large (4 mm. across eyes) ; frontal impressions light but distinct; eyes 
convex. Prothorax broader than long (4.3 x 5.4 mm.), depressed, as wide at 
base as at apex (4.1 mm.) ;_ sides rounded, obliquely narrowed posteriorly, lightly 
sinuate just before base; anterior angles obtuse; base truncate, angles rectangular, 
obtuse at summit; lateral border narrow; lateral basal impressions parallel, deep, 
rather elongate and wide. Elytra truncate-oval (11.2 x 7), dehiscent at apex; 
lateral apical sinuosities well developed, wide; basal border considerably raised, 
but joining lateral border at humeral angles "without interruption; interstice 10 
rather elongate, depressed. Intercoxal declivity of presternum rounded in middle, 
of mesosternum concave. Apical ventrite in c? with one, in $ with two punctures 
on each side. Length, 18 — 20; breadth, 6 — 7,2 mm. 

Eab. — Victoria: Mounts Hotham and Feathertop. I found it plentifully 
under logs on the road to Grant (between the sources of the Dargo and Wonnan- 
gatta Rivers) at 4,700 feet, and on the slope of Mt. Feathertop (5,000 feet) in 
February. 

Very closely allied to N. muelleri SI., from which it differs chiefly by sides 
of prothorax sinuate posteriorly, (^ with apical ventrite unipunctate on each side.* 
It resembles N. peroni Cast., so closely that in my note on N. peroni (These Pro- 
ceedings, 1913, 421), when I knew only the ?, I took it for a form of that species 
and recorded it as "N. peroni, var. D. Black." Now that the <S is known it is seen 
that in addition to its black colour it differs from N. peroni by d" with joints 1 — 3 
of the anterior tarsi dilatate and squamose beneath, 

* A^. nme/leri has sometimes a faint greenish tinge on the prothorax and elytra. 
Of seven male specimens which I have examined, five had at each side of the 
apex of the abdomen two setigerous punctures near together, and two had two 
setae on one side, but only one on the other. Only two out of nineteen male 
specimens of A', dehiscens examined, were found to have two setigerous punctures 
on one side of apex of the abdomen, all the others having one seta on each side. 
Collecting between Mts. Kosciusko and Hotham may produce intermediate forms 
showing N. muelleri and A'', dehiscens to be forms of one species; but, even then, 
I believe a name for each form will be convenient. 



BY T. G. SLOANE. 25 

NoTONOMUs HOPSONi^ n.sp. 

Elongate-oval, convex. Head stout; prothorax cordate, posterior angles not 
marked, posterior marginal seta distant from base; elytra oval, deeply striate, 
interstices convex, 3 quadripunctate, 8 narrow, lateral apical sinuosities deep, in 
<S roundly curved, in ? semicircular, inflexed margin truncated by apical sinuosities 
(widely truncated in ?). Black; elytra varying from an obscure bronze colour 
to bright-cupreous, often dull purple in c?. 

Head rather large, convex (3.5 mm. across eyes). Prothorax broader than 
long (4 X 4.4 mm.), widest before middle, narrower at base (2.8 mm.) than apex 
(3.8 mm.) ; sides rounded, obliquely narrowed to base, sometimes subsinuate near 
base ; basal angles obtuse ; lateral border narrow ; lateral basal impressions shallow, 
wide. Elytra oval (10.3 x 6 mm.), convex, humeral angles rounded; basal border 
well developed, not dentate, but a little raised above lateral border at shoulders. 
Intercoxal declivity of prosternum rounded, of mesosternum very lightly concave. 
Length, 16 — 22; breadth, 5.1 — 5.4 mm. 

Hah.—N.S. "Wales: Eccleston. I found this species generally on the AUyn 
River below the level of 2,600 feet in March, 1921. It is the species I have er- 
roneously recorded as N. johnstoni (These Proceedings, 1916, 197) from the lower 
brushes of the Williams River. 

This species belongs to the N. excisipennis-group, but differs from the other 
two members of that group by having the lateral apical sinuosities of the elytra 
more strongly excised. Comparing the 5 (no specimen of iV. johnstoni c? is avail- 
able), the differences of the apical sinuosities readily distinguish it from N. 
johnstoni', in N. johnstoni these sinuosities sharply abbreviate the inflexed margin, 
and are strongly excised, but their backward course is a gentle curve, in which 
the internal plica is seen as a narrow process; in N. hopsoni the inflexed margin 
is wider at apex, so that it is more decidedly abbreviated, and, owing to the 
strong outward dilatation of the inner plica, the lateral sinuosity is deeper and 
forms a more or less semi-circular notch. The apical sinuosity in c? is almost as 
in N. johnstoni 9, but the margin of the elytra round the sinuosity is more curved. 
The raised process of the apical ventral segment which fits into the apical 
sinuosity is considerably more developed in N. hopsoni than in N. johnstoni. 

NOTONOMUS ELLIPTICUS, n.Sp. 

?. Elongate. Head rather large (3.3 mm. across eyes); prothorax sub- 
quadrate, depressed, posterior marginal puncture on inner side of marginal 
channel near basal angle; elytra deeply striate, interstices convex, 3 bi-punetate, 
8 narrow. Nitid, upper surface (including head) cupreous, under surface and 
legs black; tarsi and antennae reddish-piceous. 

Eyes convex, prominent; frontal impressions deep, short, divergent back- 
wards. Prothorax broader than long (4.2 x 4.55 mm.), lightly rounded on sides, 
wider across base (3.5 mm.) than apex (3.3 mm.) ; base emarginate in middle, 
basal angles marked, obtuse at summit; border strongly reflexed, subsinuate just 
before base, extending alongvbase on each side to lateral basal impressions; these 
impressions narrow, deep; median line strongly impressed. Elytra oval (10 x 5.5 
mm.), depressed on disc; apical curve strongly sinuate on each side; basal border 
strongly raised and prominent at humeral angles; striae deep, simple; interstices 
convex, narrow and earinate near apex, 8 narrow, wider than 9 on basal half, 9 



26 STUDIES IN AUSTRALIAN ENTOMOLOGY^ xviii., 

very narrow, 10 well developed, extending forward from apical sinuosities half 
way to base. Intercoxal declivity of prosternum flat, of mesosternum lightly 
concave. Length, 18; breadth, 5.5 mm. 

Hab. — Queensland. Two specimens (?) in British Museum ticketed "Queens- 
land, Challenger Exp." are the types of this species. I have seen a specimen 
from Booyong, Richmond River, N.S. "Wales. 

It belongs to the N. opuicicolUs-gron'p of the genus, being allied to N. wilcoxi 
Cast., from which it differs by colour more metallic (including head) ; size larger; 
frontal impressions of head deeper; prothorax less narrowed to base; elytral 
interstices more convex, particularly on apical declivity. 

NOTONOMUS CURVICOLLIS^ n.Sp. 

c?. Elongate-oval, convex. Head stout; prothorax transverse, strongly and 
evenly rounded at sides, posterior marginal seta on border at basal angles; elytra 
oval, fully and strongly striate, interstice 3 quadri-punctate on posterior two- 
thirds. Joint 1 of intermediate tarsi 2-spinulose, of posterior 1-spinulose be- 
neath eosta of outer side. Elytra vireseent, prothorax viridaeneous, head aeneous 
on front and vertex ; under surface and legs . black ; apex of tibiae, tarsi, and an- 
tennae somewhat reddish. 

Head convex (3.5 mm. across eyes); eyes roundly convex. Prothorax 
broader than long (4.4 x 5 mm.), convex; apex and base of equal width (3.35 
mm. ) ; sides evenly arcuate ; basal angles roundly obtuse ; lateral border wide, 
reflexed; lateral basal impressions short, oval, deep. Elytra oval (11 x 6.2 mm,), 
conve?.; ante-apical sinuosities shallow, oval; interstices hardly convex, 8 wider 
than 9 on basal half, 10 well developed, not long. Intercoxal declivity of pros- 
ternum rounded and narrow in middle, of mesosternum widely concave. Length, 
18.5; breadth, 6.2 mm. 

Hab. — Victoria. Unique in Coll. Sloane. I found it in February on the 
steep side of Mount Feathertop under felled timber at 5,700 feet. 

AUied to N. rainbowi SI., N. hanksi SI., and N. aequalis SI. It is at once 
distinguished from N. rainhowi by smaller size and green colour. From N. 
hanksi it differs by smaller size, prothorax shorter and more strongly rounded on 
sides. It is about the same size as N. aequalis, but differs by prothorax more 
strongly rounded at sides, interstices of elytra less convex, particularly towards 
apex, etc. 

Tribe Merlzo'dmi. 

Genus P T e r o c y r t u s. 

Pterocyrtus truncaticollis, n.sp. 

Stout, elongate-oval. Head stout; prothorax subquadrate, basal angles rect- 
angular; elytra oval, punctate-striate, basal border well developed. Black, border 
of prothorax and elytra dull red; antennae, mouth parts, and legs reddish; 
femora reddish piceous. 

Head convex (0.9 mm. across eyes); frontal impressions short, wide; clypeus 
with a foveiform puncture on eaeli side; eyes round, convex, but not prominent, 
distant from buccal fissure beneath; 2 supra-orbital setae on each side. Pro- 
thorax subquadrate (1 x 1.35 mm.), broadest before middle, Avider across base 
than apex; sides hghtly rounded anteriorly, lightly oblique posteriorly; anterior 



BY T. G. SLOAXE. 27 

angles obtuse, a little advanced; base wide, truncatej basal angles rectangular 
but obtuse at summit; border narrow, not extending on to base; lateral basal 
impressions deep, wide, short, punetulate; space between basal impressions and 
lateral channel narrow, convex; posterior marginal seta at basal angle. Elytra 
oval (2.5 X 1.8 mm.), convex, fully striate; striae punctate; diseal interstices a 
little convex, 1 with a short basal striole, 3 tri-punctate, 8 carinate posteriorly. 
Length, 4.3; breadth, 1.8 mm. 

Hab. — Victoria. I obtained three specimens in a dry situation, under moss 
on a rock, beside the road from St. Bernard Hospice to Grant in the mountains 
at the source of the Dargo and Wonnangatta Rivers, at 5,000 feet. 

I have referred this species to the genus Pterocyrtus with doubt; it is the 
first species from the mainland to be put in this genus, with which it agrees in 
all characters, except that the head has two supra-orbital setae on each side; the 
elj^tra have the basal border strongly developed and reaching to the first inter- 
stice, raised but not dentate at humeral angles; there is a short seutellar striole at 
base of first interstice. It will form the type of a distinct section in the genus. 
Its narrow subquadrate prothorax and punctate-striate elytra give it quite a 
different faeies from Bracliydema tasmaniae SI. 

Tribe Bemibidiini. 

Genus T A c h y s. 

Tachys sydneyensis_, n.sp. 

Oval, convex, laevigate. Two minute setiferous punctures on disc of each 
elytron. Piceous; legs dull testaceous; antennae infuscate, basal joint testaceous. 

Head wide, convex; frontal impressions wide apart, shallow, short, parallel. 
Prothorax transverse, truncate-cordate; sides rounded anteriorly, subsinuate be- 
fore base; basal angles subrectangular, obtuse at summit; base wide, transversely 
impressed in middle near edge; lateral basal impressions obsolete; border very 
narrow, extending on to base on each side. Elytra oval, convex, laevigate, apical 
striole very short and feebly impressed, three marginal punctures behind shoulders 
and three on apical curve. Length, 1.4; breadth, 0.6 mm. 

Hah. — N.S. Wales: Sydney. Type in Coll. Sloane. 

This species was obtained by me in December on the edge of a small pool 
among the sandy hills near Manly. Two specimens from the Castelnau collection 
in the Museum of Genoa were sent to me by Dr. Netolitzky, to whom they were 
returned under the name of ''T. sydneyensis (cotypes)." 

Its nearest allies are T. austr aliens SI., and T. captus Blaekb. (also Tachyta 
Uvida Bates — unknown to me in nature) ; from all of these it differs in the com- 
plete absence of striae on the elytra, and from T. captus and T. Uvida by its 
convex form; in this it resembles T. australicus, but differs by the want of the 
sutural stria; in T. australicus, too, the prothorax has the basal angles more 
acute, and the lateral channel narrower beside these angles. 

Tribe Trechini. 

Genus T R E c H u s. 

Trechus kosciuskoanus^ n.sp. 

Oval, rather depressed. Prothorax transverse, much wider across base than 

apex, base truncate, basal angles rectangular; elytra finely striate, apical striolae 



28 STUDIES IN AUSTRALIAN ENTOMOLOGY^ Xviii., 

distinct; interstice 3 tri-punctate, two small diseal punctures beside stria 3, pos- 
terior puncture on apical declivity, basal border extending inwards to base of 
interstice 3. Black, nitid; femora pieeous, tibiae and tarsi ferruginous; antennae 
piceous, basal joint ferruginous; palpi pieeous. 

Head large (1 mm. across eyes) ; frontal sulci curved, strongly impressed 
but not deep; eyes convex, but not prominent; post-ocular part of orbits small, 
projecting a little from head. Protborax broad (1 x 1.3 mm.); anterior angles 
wide, rounded; sides ligbtly arcuate anteriorly, bardly narrowed to base; lateral 
border narrow; lateral basal foveae wide, deep. Elytra broad (2.5 x 1.8 mm.), 
rather depressed on disc, a little declivous to base; humeral angles marked, and 
with border strongly raised; base wide, four inner striae distinct, 5 — 7 faint but 
perceptible. Length, 3.8 — 4.5; breadth, 1.65 — 1.8 mm. 

Hab. — N.S. Wales: Mount Kosciusko. Several specimens collected by me on 
7th December under stones beside the track to the summit at from 6,800 to 7,300 
feet. 

A distinct species more allied to T. robiisfus SI., than to any other, but, 
comparing it with the description of that species, its smaller size and shining 
black colour are in themselves sufficient to distinguish it. From T. diemenensis 
Bates it is differentiated by colour; prothorax more transverse, much wider across 
base, more arcuate on sides, lateral border narrower and less upturned near base; 
elytra with striae finer, apical striolae not hooked, punctures of third interstice 
smaller, the anterior puncture beside third stria not interrupting the interstice. 

Teechus australiensis, n.sp. 

Elliptical-oval, depressed. Prothorax transversely truncate-cordate, basal 
angles rectangular; elytra finely striate, apical striolae distinct, interstice 3 tri- 
punctate, two small diseal punctures beside stria 3, posterior puncture on apical 
declivity, basal border extending inward to stria 4. Reddish brown, head and 
prothorax more reddish than elytra, interstice 1 of elytra usually reddish; legs 
and antennae ferruginous. 

Head large (0.7 mm. across eyes); frontal impressions curved, deep, de- 
fining post-ocular part of orbits; eyes convex, but not prominent; post-ocular 
part of orbits protuberant. Prothorax broader than long (0.65 x 0.85 mm.); 
base truncate, a little narrower than apex; lateral border rather wide; lateral 
basal impressions narrow, rather deep. Elytra oval (1.7 x 1.3 mm.), depressed, 
a little declivous to base; humeral angles rounded; border narrow, three inner 
striae distinct, 5 and 6 faint. Length, 2.8 — 3.2; breadth, 1.2 — 1.3 mm. 

ITab.— N.S. Wales: Mount Kosciusko. I collected it on 7th December not 
uncommonly under stones beside the track to the summit at from 6,800 to 7,300 
feet. 

More nearly allied to T. kosciushoanus SI. than to any other species, but 
differently coloured, much smaller, proportionately narrower; prothorax more 
narrowed to base, which is not wider than apex. Compared with T. diemenensis 
Bates, it is smaller; prothorax with lateral border less reflexed near ba.se; elytra 
less strongly striate, anterior puncture of third interstice beside third stria and 
not interrupting the interstice. 



BY T. G. SLOANE. 29 

Trechus gippslandicus^ n.sp. 

Oval, convex. Elytra much wider than prothorax. Piceous-blackj elytra 
variegated (pattern testaceous), inflexed margin testaceous; femora and base of 
antennae light coloured. 

Head elongate, narrow; frontal impressions deep, curving outwards an- 
teriorly and posteriorly. Prothorax narrow in proportion to elytra, broader than 
long (1.5 X 1.65 mm.), widest before middle, hardly narrowed behind, wider 
across base than apex, subdepressed on disc, roundly declivous to sides; basal 
angles marked, obtuse at summit; base rounded in middle, sloping obliquely in- 
ward from each side; marginal channel wide; m£>j:gin wide, explanate and strongly 
reflexed at basal angles; basal foveas deep, wide, reaching to lateral channel; 
median line lightly impressed. Elytra oval (3.6 x 2.5 mm.), convex, lightly 
depressed on disc, strongly roundly declivous to sides and apex, rounded at sides; 
shoulders rounded; a light sinuosity at each side of apex; six inner striae strongly 
impressed, simple, 7 and 8 obsolete (except 8 towards apex) ; stria 1 entire, 
curving round apex and extending forward in a short deep course opposite ex- 
tremity of 5; interstices depressed, 3 tri-punctate (two anterior punctures on 
disc near stria 3, posterior one at extremity of 2) ; marginal channel narrow 
along side, explanate just before apical sinuosity. Length, 5.5; breadth, 2.5 mm. 

Hah. — Victoria. Unique in my collection, received from Mr. C. French as 
from Gippsland. 

Allied to T. suhornatelltis Blackb., but larger and with a different elytral 
pattern. The ground colour of the elytra is piceous; there is a narrow light 
coloured margin (including interstice 1 towards apex) ; a testaceous mark ex- 
tends from a little before the middle of interstice 4 obliquely forward to the 
side of each elytron, and another similar mark extends across interstices 4 — 7 a 
little behind the level of the posterior discal puncture; these two marks unite on 
interstice 7; interstice 3 has a small plaga in front of the posterior discal punc- 
ture. In T. suhornatellus the testaceous markings form a wavy broken fascia 
across the elytra above the apical declivity, and there are no testaceous marks 
on the basal half of the elytra. 

Trechus bitinctus_, n.sp. 

Depressed. Head arcuately bisulcate, eyes hemispherical; prothorax trans- 
versely subcordate, base lobate as in T. hipartitus MacL; elytra truncate-oval, 
iinistriate on each side of suture, each elytron 3-punctate (two discal punctures 
strongly marked), scutellar striole obsolete, basal border reaching seutellum; 
abdomen setulose. Upper surface bicoloured; head, prothorax, base and sides 
of elytra to level with posterior discal puncture rufescent, apex and disc of elytra 
almost to anterior puncture shining black, with a virescent tinge in some lights; 
under surface ferruginous, becoming of a lighter shade on abdomen; antennae 
and legs testaceous. 

Head large (0.9 mm. across eyes); frontal sulci deep, curved, strongly 
divergent posteriorly; median frontal space subconvex, wider than lateral spaces, 
rather convex. Prothorax rather convex, broader than long (0.8 x 1 mm.), a 
little wider across basal angles than across apex; sides strongly rounded anteriorly, 
gently narrowed behind; basal angles prominent, shortly subdentate with apex 
blunt; basal curve oblique behind basal angles, bisinuate in a well marked open 



■30 STUDIES IN" AUSTRALIAN ENTOMOLOGY^ xviii., 

curve on each side of median lobe; posterior margin of lobe arcuate; anterior 
transverse impression deep and curved backwards; posterior transverse im- 
pression well marked between basal sinuosities; lateral border reflexed. Elytra 
smooth, suboval (2.1 x 1.5 mm.); base wide; sides subparallel; apical curve 
short; submarginal ^ stria (8) distinct behind humeral angles, its course inter- 
ruptedly indicated on posterior third. Length, 3.5; breadth, 1.5 mm. 

Hob. — Northern Territory: Adelaide River. 

Three specimens have been examined, one belonging to the collection of the 
British Museum (type, ticketed "Adelaide R., 91 — 49"), and two others (one in 
Coll. Sioane) found by Mr. H. W. Brown on the Adelaide River. Closely allied 
to T. tipartitus Mael., from which it differs by its bicoloured elytra. The an- 
terior tarsi in c? have the two basal joints dilated with the outer angles dentate. 

Tribe Odacanthimi. 

Six years ago I reviewed the tribe Odacanthini, as represented in the Aus- 
tralian fauna (These Proceedings, 1917, 413). Since then I have obtained a new 
and remarkable species, for which I believe a new genus is required (c/. Aula- 
coliris, postea), and, my views on the generic distribution of our species having 
undergone considerable modification, I now offer a revised tabulation of our 
genera, which I hope will enable them to be recognised. Some notes are added 
after the table to show my altered views with regard to some of the genera. 

Table of Genera. 

1 (2) Head not constricted to a neck; eyes at a normal distance from prothorax. 

Porocara (1917). 

2 (1) Head narrowed behind eyes, neck (except rarely) narrow and condyli- 

form; eyes distant from prothorax. 

3 (14) Antennae with joints 1-3 glabrous. Tarsi glabrous on upper surface. 

4 (13) Elytra with punctate striae. 

5 (12) Antennae with joints 3 and 4 subequal. 

6 (11) Tarsi with joint 4 entire. 

7 (10) Prothorax not cylindrical, hardly narrower at apex than base, lateral 

border always present and reaching base. 

8 (9) Elytra fully striate Dicraspeda (1862). 

9 (8) Elytra striate on basal fourth, laevigata on apical three-fourths 

Basistichus (1917). 

10 (7) Prothorax cylindrical, much narrower at apex than base, lateral border 

short, not reaching base. . Arame (1919). 

11 (6) Tarsi with joint 4 bifid Oplvionea (1821). 

12 (5) Antennae with joint 3 elongate (about as long as 4 and 5 together) 

. Clareneia (1917). 

13 (4) Elytra fully striate, striae simple Aulacolkis (n.g.) 

14 (3) Antennae with joint 3 setulose. Tarsi setulose on upper surface 

Myrmecodemus (n.g.) 
I do not know Castelnau's genus Anasis in nature. 

Genus Dicraspeda. 

Cbaudoir, Bull. Soe. Imp. Nat. Mosc, 1862, 300. EudaUa, Castelnau, Trans. 
Roy. Soc. Vict., viii., 1868, 102; Bates, Ent. Mo. Mag., 1871, viii., 32. 

Formerly, I had not taken the genus Dicraspeda into account, but now feel 
compelled to regard it as synonymous with Eudalia, which must be merged with 



BY T. G. SLOAjSTE. 31 

it. E. sublaevis Macl. agrees so well with the description of D. hrunnea Chaud., 
from Celebes, that I believe it to be the same thing; there seem no reasons for 
referring E. sublaevis to a diiferent genus from Oclacantha latipennis Macl., the 
genotype of Eudalia. In These Prcceedings, 1917, 414, I gave a synoptic table 
of the species of Eudalia, as then understood by me; all these species may be 
transferred to Dieraspeda, except Ctt'Snonia obscura Cast., which I would place 
in Arame. 

Genus A b a m e. 

The old magazine-genus Casnonia has been shown by Mr. H. E. Andrewes to 
be invalid (Ann. Mag. Nat. Hist., (9), iii., 1919, 477) ; at the same time he 
founded a new genus Aramej for some Oriental species which had formerly been 
placed in Casnonia. This genus Arame will include Casnonia obscura Cast. 

Genus p h i o n e a. 

The genus Ophionea has been given a narrow significance by being confined 
to those species which have the fourth joint of the tarsi bifid, but it seems very 
doubtful whether this character should be allowed to restrict the limits of this 
genus, though this is a matter that does not concern the Australian fauna, the 
two known Australian species of Ophionea having the fovirth tarsal joint bifid. 

Ophionea puncticollis^ n.sp. 

Elongate. Prothorax elliptical, subcylindrical, lateral border indicated, pro- 
notum and prosternum punctate; elytra with two white ante-apical spots, punc- 
tate-striate, interstices depressed; joint 5 of tarsi deeply bifid. Head nigro- 
eyaneous; prothorax and base and apex of elytra reddish, a wide transverse 
dark cyaneous band forward from white maculae; legs testaceous, apical third 
of femora, extreme apex of tibiae, and apices of tarsal joints infuscate; antennae 
infuseate, joints 1 — 4 testaceous. 

Head ant -like, convex (1.4 mm. across eyes), roundly-obliquely and rather 
shortly narrowed to the condyliform neck; frontal sulci wide and distinct be- 
tween antennae, diverging near eyes and extending to basal third of eyes ; a sub- 
marginal costa between posterior part of each frontal sulcus and eye. Prothorax 
narrow (1.7 x 1 mm.), wider at base than apex, transversely impressed near 
base; sides lightly arcuate in middle, lightly narrowed to apex, shortly sub- 
rectangular near base; upper surface punctate and transversely rugose, punctures 
and rugae more distinct towards sides; apex with a strong narrow border. Elytra 
parallel (4x2 mm.). Lengih, 8; breadth, 2 mm. 

Hab. — Queensland: Burjlekin River (type, in Coll. Sloane) ; Northern Terri- 
tory: Darwin (G. F. Hill). 

Compared with 0. thouzeti Cast., it differs by colour (base of elytra not 
cyaneous), head more obliquely narrowed behind eyes, prothorax punctate and 
with lateral borders developed on anterior half. Compared with the Oriental 
species known to me (e.^/,. 0. cyaneocephala Fabr.), the head is stouter, more 
shortly and less obliquely narrowed behind eyes, prothorax less rounded at widest 
part, much less narrowed to apex, etc. 



32 STUDIES IN AUSTRALIAN ENTOMOLOGY^ xviii., 

Genus Aulacolius^ n.g. 

Head glabrous, diamond-shaped, constricted by an elongate slope to a narrow 
condyliform neck; antennae with joints 1 — 3 glabrous, joints 3 and 4 subequal. 
Prothorax oviform, impunctate — except on the naiTow basal part; pronotum 
gibbous, transversely striolate, impunctate; lateral borders well developed, not 
reaching base; lateral channel with four or five widely spaced setae on the am- 
pliate part of sides. Prosternum convex, sides visible from above. Elytra 
ovate, twice as wide as prothorax, fully striate ; disc transversely impressed across 
interstices 2 — 4; striae entire, simple; interstices equal, 3, 5 and 7 seriately pluri- 
punetate, the punctures bearing rather long setae, and placed in two rows on 
outer sides of interstices 3 and 5; four punctures on interstice 1 beside scutellar 
striole. Tarsi glabrous on upper surface, joint 4 entire. 

I do not know any genus to which the species described below as Aulacolim 
triordinatus can be referred. It cannot be placed far from Lachnothorax, but is 
distinct by pronotum and striae of elytra impunctate, head and pronotum without 
setae, except the fixed ones. 

AULACOLIUS TRIORDINATUS, n.Sp. 

$. Head diamond-shaped, neck narrow, condyliform; prothorax oviform, 
longer than broad, ampliate on sides, impunctate except near base, lateral borders 
well developed; elytra wide, disc transversely impressed near basal third, striae 
entire, deep, simple, interstices depressed, 3, 5 and 7 pluripunetate, others laevi- 
gate. Black; apex and apical half of inflexed margins testaceous; antennae 
piceous, basB more lightly coloured; legs testaceous, coxae and posterior trochan- 
ters clear brown, femora at apex, tibiae at base, and apex infuseate ; tarsi in- 
fuseate; abdomen with segments 5 and 6, also sides of 3 and 4 clear brown. 

Head convex, impunctate, glabrous (1.4 mm. across eyes), strongly obliquely 
angustate to base; eyes convex, prominent. Prothorax a little narrower than 
head (1.6 x 1.2 mm.), sides roundly ampliate in middle, decidedly rounded an- 
teriorly, strongly narrowed posteriorly, strongly sinuate at posterior fifth; basal 
part rugose-punctate, defined by a strong transverse impression, lateral border 
distinct, except behind transverse basal impression, lateral channel well developed 
and bearing four or five setae on ampliate part of sides; pronotum gibbous, 
transversely striolate. Elytra truncate-oval (3.3 x 2.6 mm.), declivous to 
peduncle; apex wide, apical curve short, a little oblique outwards from inter- 
stice 4; striae strongly impressed, smooth; an elongate impunctate striole at base 
of interstice 1; interstices depressed, inner ones a little raised towards apex, odd 
ones seriately setigero-punctate, even ones nitid, impunctate; four or five seti- 
gerous punctures along outer side of scutellar striole; lateral channel impunc- 
tate, intumed to meet stria 8 near base; inflexed margins glabrous, impunctate. 
Under surface impunctate, except peduncle and base of prothorax. Length, 8; 
breadth, 2.6 mm. 

TIab. — Northern Territory: Darwin. Unique in Coll. Sloane. Mr. G-erald 
P. Hill found one specimen which he generously presented to me. 

This species is at once differentiated from all our other Odaeanthides by the 
striae of the elytra being entire and impunctate. 



BY T. G. SLOANE. 30 

Myrmecodbmus-, n.g. 

Head strongly constricted in a curve to a short condyliform neck; antennae 
with joint 3 about equal in length to 4, sparsely setulose. Prothorax oviform; 
base narrow, defined by an encircling sulcus, pronotum gibbous, sparsely setose, 
bordered on sides ; lateral channel with several setae at ampliate part of sides. 
Prosternum convex, smooth; sides visible from above. Elytra ovate, about twice 
as wide as prothorax, transversely impressed at basal third, sparsely seriate- 
setose. Tarsi sparsely setose on upper surface, joint 4 entire. 

Genotype, Casnonia riverinae SI. 

This new genus is allied to Selina, but has the neck short (not forming a very 
narrow stalk to the head) and the third joint of the antennae not elongate. It is 
founded on Casnonia riverinae SI., to which C. globuUcolUs Macl., is closely al- 
lied, also Lachnotkorax formicoides SI. LacJinothorax palustris SI., is very dis- 
tinct from the three species mentioned, and perhaps is not actually congeneric 
with them, but it is more in place in the gen vis Myrmecodemus than anywhere 
else. I was mistaken in referring these four species to the genus LacJino- 
thorax in 1917, but the tabulation then proposed will enable them to be recog- 
nised. The typical species of LacJinothorax have joint 3 of the antennae glabrous. 

Tribe Panagemi. 

Genus Ceaspedophorus. 

Hope, Col. Man., ii., 1838, 165; Chaudoir, Ajlu. Soc. Ent. Belg., xxi., 3878, 
90.— Eudema Castelnau, Hist. Nat. Ins., I., 1840, 137; Trans. Roy. Soe. Yict., 
viii., 1868, 145; Chaudoir,J.c., 133. — Epicosmus Chaudoir, Bull. Soe. Imp. Nat. 
Mosc, iv., 1846, 512; Ann. Soe. Ent. Belg., xxi., 104; Sloane, Proc. Linn. Soc. 
N.S.W., 1903, 566. 

In his Essai MonograpJiique sur les Panageides (1878) Chaudoir used the 
three names CraspedopJiorus, Eudema and Epicosmus as being each entitled to 
rank as a distinct genus, and in doing so he used CraspedopJiorus rightly, but it 
is impossible to follow him in his ideas as to the recognition and use of Eudema 
and Epicosmus. The facts are, as has been explained by Mr. H. E. Andrewes 
(Trans. Ent. Soc. Lond., 1919, 126), that Eudema Castelnau is synonymous with 
Craspedophorus Hope, while the type of Epicosmus is Carahus angulatus Fabr., 
which is erroneously referred to under the name of Eudema bifasciatum Fabr. (a 
name never used by Fabrieius) in Chaudoir's monograph, where it forms the only 
species of the genus Eudema. I support Mr. Andrewes's view that both Eudema 
and Epicosmus should be sunk as synonyms of Craspedophorus. 

In These Proceedings for 1903, 566, under Epicosmus, I dealt with our 
species, giving a table of species and proposing three new specific names, but I 
had too little data then, and now see that my E. ohesulus is conspecific with E. 
rocJchamptonensis Cast., and E. froggatti with E. australasiae Chaud. I now 
have two new species to describe, and give below a new table of species leading 
off from Cr. parvulus Macl., which is the most nearly related of our species to 
those of Java and Borneo. It seems to me certain that the Australian Panageini 
have been derived from the Oriental region. 



\ V- '^ " C 



34 STUDIES IN AUSTRALIAN ENTOMOLOGY;, XViii., 

Table of Australian Species. 

Prothorax with posterior angles rectangular and dentate. 

Elytra with maculae orange, or lemon-coloured. 

Elytra somewhat lightly furrowed; interstices roundly convex, equal, 

setose-punctate. 9.5 mm parvulm Macl. 

Elytra deeply furrowed; interstices 3, 5 and 7 rather stronger than others. 
Elytra with two (rarely one) fixed punctures on interstice 3, maculae 
reaching to third stria. 

Prothorax strongly sinuate-angustate to base. 12-13 mm. comptm Laf. 
Prothorax not sinuately narrowed to base. 12.5-13.5 mm 

roekhamptonensis Cast. 
Elytra without fixed punctures on interstice 3, maculae reaching to fourth 
stria. 
Head narrow before eyes; elytral interstices evidently setose-punctate. 

11-12.5 mm angusticeps SI. 

Head stout before eyes; elytral interstices nitid, hardly punctate, 13.5- 

15.5 mm. lanksi SI. 

Elytra with maculae red. (Interstices 3, 5 and 7 more raised than 

others). 

Form oblong, subdepressed. 

Elytra with third interstice unipunctate near posterior macula. 15.5-18 

mm , . .alternans Cast. 

Elytra with third interstice impunctate. 12-14 mm mastefsi SI. 

Form oval, prothorax and elytra convex. Interstices 1 and 2 of elytra 

evidently depressed below 3. 10-12 mm australasiae Chaud. 

Prothorax with posterior angles rounded off. 

Anterior macula of elytra reaching to third stria. 17.5-21 mm 

insignis Sehaum. 
18 (17) Anterior macula of elytra reaching to fourth stria. 17 mm. macleayi SI. 

Craspedophorus angusticeps^ n.sp. 

Oval. Head narrow; prothorax small; elytra with interstices very convex, 
almost equal. Black; four citron-coloured maculae on elytra, anterior pair dis- 
tant from base, extending across interstices 5 — 9, posterior pair small, extending 
across interstices 5 — 8, uneven on anterior and posterior edges. 

Head elongate (1.6 mm. across eyes); frontal part longer than broad (1.3 
X 1.2 mm.), finely punctate, laevigate anteriorly. Prothorax broader than long 
(2.6 X 3.2 nun.), a little convex on anterior part of disc, widest behind middle, 
strongly narrowed to apex (1.6 mm.), obliquely narrowed to base (2.4 mm.); 
anterior angles narrow, obtuse; lateral margins declivous anteriorly, widely up- 
turned posteriorly; lateral border narrow; basal angles subdentate. Elytra con- 
vex, oval (6.8 X 4.7 mm.) ; interstices sparsely setose-punctate, 3 — 7 subequal, 5 
dilated and 4 narrowed at posterior macula. Length, 11 — ^12.5; breadth, 4.5 — 
5.2 mm. 

Bah. — Queensland: Mareeba (Sloane), Cooktown (Brown), Coen (Hacker). 
Tj-pe in Coll. Sloane. Twelve specimens have been examined, most of which were 
found by me under logs beside the railway station at Mareeba in June. 

This species is characterised by its small size, narrow head, interstice 3 of 
elytra without fixed punctures, maculae of elytra pale yellow; the anterior 
macula is transverse, and consists of five spots across interstices 5 to 9, the spots 



1 

2 
3 


(16) 

(11) 

(4) 


4 
5 


(3) 
(8) 


6 

7 


(7) 
(6) 


8 


(5) 


9 


(10) 


10 


(9) 


11 


(2) 


12 


(15) 


13 


(14) 


14 


(13) 


15 


(12) 


16 


(1) 


17 


(18) 



BY T. G. SLOANE. 35 

on interstices 8 and 9 are longer than the three inner spots; the posterior macula 
is smaller, and consists of four spots across interstices 5 — 8, the spot on inter- 
stice 6 shortest; the inflexed margins usually have a faint reddish mark near the 
edge below the anterior elytral macula, but sometimes this mark is absent. The 
interstices of the elytra are more setose than usual, the setae rising from small 
punctures on the interstices, but not from the large punctures of the striae. 

Craspedophorus banksi^ n.sp. 

Elongate-oval, subdepressed. Head wide; prothorax transverse, strongly 
punctate, subobliquely (roundly) narrowed to base; elytra with interstices sub- 
equal. Black; four citron-coloured maculae on elytra, anterior pair extending 
across interstices 5 — 9, posterior pair narrow, extending across interstices 5 — 8, 
anterior and posterior margins uneven, spots on interstices 6 and 8 shorter than 
those on 5 and 7, inflexed margin with a reddish mark below anterior macula. 

Head stout (2.45 mm. across eyes); frontal part wide (1.6 x 1.7 mm.); 
frontal sulci punctate; interspace convex, laevigate anteriorly, punctate pos- 
teriorly between eyes; antennae elongate. Prothorax transverse (3.5 x 4.5), 
widest behind middle, lightly convex (but not roundty raised) in middle of disc; 
sides strongly and roundly narrowed to apex (2.4 mm.), subobliquely and rather 
roundly narrowed to base (3 mm.); anterior angles rounded; basal angles feebly 
dentate; median line well marked; lateral basal impressions long, sulciform, 
placed opposite stria 4 of elytra. Elytra ovate (9 x 5.65 mm.), lightly rounded 
at sides, lightly sinuate on each side of apex; disc depressed between third inter- 
stices; striae very deep, crenulate; interstices 1 — 8 strongly convex, their summits 
nitid, 3 without fixed setiferous punctures, 3, 5, and 7 wider than others, 4 de- 
cidedly narrowed at posterior macula, 5 swollen on inner side at posterior macula. 
Length, 13.5 — 15.5; breadth, 5.4r— 6.2 mm. 

Hah. — Queensland: Cooktown and Upper Normanby River (Sloane), Kur- 
anda (Dodd), Cairns. Type in Coll. Sloane. I found several specimens under 
Eucaljrptus logs on the sides of Mount Cook at Cooktown, and at King's Plains 
cattle station on the Normanby River. 

A large species which differs from Cr. alternans Cast., by its shorter and 
wider head; shorter and wider prothorax, not sinuate on sides posteriorly; eljrtra 
with the maculae not extending on to interstice 4, interstice 5 swollen and 4 
angustate at posterior macula, the alternate interstices not so much raised above 
the others. It is very like Cr. roclchamptonensis Cast., which it resembles in the 
colour of the maculae of the elytra, but it differs by antennae longer, the joints 
after 3 longer and slenderer; prothorax less strongly declivous to sides before 
middle, lateral margins wider; maculae of elytra not extending on to interstice 4, 
the posterior macula smaller, more uneven on anterior and posterior margins, the 
spot on interstice 6 shortest, interstice 5 swollen and 4 angustate at posterior 
macula. 

Tribe Licinini. 

Genus Diorochile. 

Table of Australian Species. 

1 (4) Elytra with third interstice 3-punctate. 

2 (3) 5. Elytra nitid. .. •• •- . B. quadrieolUs Cast. 

3 (2) $. Elytra opaque, shagreened , B. ventralis Blackb. 

4 (1) Elytra with third interstice 2-punctate. 



56 STUDIES IN AUSTRALIAN ENTOMOLOGY^ XViii., 

5 (6) Striae of elytra deep, crenulate D. goryi Boisd. 

6 (5) Striae of elytra simple. 

7 (12) Interstices of elytra not punctulate. 

8 (11) Size moderate. 12 mm. 

9 (10) Prothorax transverse (2.5 x 3.4 mm.) ; striae of elytra shallow, interstices 

fiat, nitid in <?, opaque in 5 B. minuta Cast. 

10 (9) Prothorax quadrate (2.3 x 3 mm.) ; striae of elytra deep, interstices 

convex, nitid in both sexes D. brevicollis Chaud. 

11 (8) Size large ; elytra nitid in ?. 21 mm. D. gigas Cast. 

12 (7) Interstices of elytra punctulate . . . . . . ■ . D. punctulata SI. 

I have not been able to identify D. montana Cast., B. punctatostriata Cast., 
and B. puncUpennis Cast. 

DiCROCHILE PUNCTULATA^ n.sp. 

V. Elliptical, depressed. Prothorax transversely cordate; elytra lightly 
striate, interstices depressed, punctulate, 3 bi-punctate beside stria 2. Black, 
nitid. 

Head transverse-quadrate (2.75 mm. across eyes); front concave; eyes pro- 
minent, hemispherical. Prothorax transverse (2.5 x 3.6 mm.), wider across apex 
(2.75 mm.) than base (2.63 mm.J, nitid on disc, a little rough and minutely 
punctulate towards sides; margins wide, reflexed (strongly so jposteriorly) ; 
lateral basal impressions wide; basal angles obtuse. Elytra parallel-oval (9 x 5.5 
mm. ) ; striae fine, subcrenujate ; interstices flat, rather closely covered with small 
punctures, these punctures more numerous on even than on odd interstices. 
Ventral segments 3 — 5 setigero-punctate on each side of middle. Length, 14.5; 
breadth, 5.5 mm. 

Hah. — Northern Territory: Darwin. Unique in Coll. Sloane. Kindly given 
to me by Mr. G. F. Hill; I have seen other specimens from East Alligator River 
in the National Museum, Melbourne. 

It is distinguished from all other Australian species of the genus by the 
punctulate interstices of the elytra; the puncturation is denser (about 4 rows) 
on the even interstices; on the odd interstices it has a tendency to be arranged in 
two rows, one along each side of the interstice. Its affinity seems to be with 
B. gigas Cast., which it resembles in shape of head and prothorax. 

Tribe Lebiilni. 
Xanthophoea wilsoni, n.sp. 

Narrow, elongate. Prothorax sinuate posteriorly, basal angles rectangular; 
elytra lightly narrowed to base; third interstice 3-punctate. Elytra brown, inter- 
stices 2, 4 and 6, and lateral margin of a dull testaceous colour, brownish near 
apex; prothorax testaceous at sides, brownish on disc; head testaceous, some- 
times rather brownish on vertex; under surface testaceous, abdomen brownish; 
legs and antennae testaceo.ys. 

Head convex (1.15 mm. across eyes), obliquely narrowed behind eyes, setu- 
lose on vertex and beside eyes. Prothorax broader than long (1.1 x 1.4 mm.), 
depressed and explanate at sides, lightly convex on disc, sparsely covered with 
small setules arising from very small punctures; sides lightly rounded on anterior 
two-thirds, sinuate posteriorly and meeting base at right angles; basal angles 
rectangular. Elytra subparallel, widest about apical fourth (3.5 x 2 mm.), sub- 
depressed on disc, strongly decjj^yous at sides from interstice 7, lightly declivous 



BY T. G. SLOANE. 37 

to base and apex; striae deep, finely crenulate; interstices subconvex, minutely, 
but sparsely, setulose punctate, 3 with three strong punctures, 5 and 7 with some 
punctures stronger than others along outer side. Under surface (including pros- 
termim and metasternum) sparsely setulose. ^ with one, $ with two setae at 
each side of apex of abdomen. Length, 5.7 — 6.5; breadth, 2 mm. 

Hob. — Victoria: Mount Donna Buang, 4,080 feet. Colls. Wilson and Sloane. 
Mr. F. E. Wilson sent me four specimens taken hj him in tussocks of grass on 
Mt. Donna Buang, near Warburton. 

This is a very distinct species, more allied to X. infuscata Chaud. than to any 
other described species, but easily distinguished by size smaller; head more con- 
vex, neck thicker; prothorax somewhat similar in shape, but with basal angles 
less prominent; elytra differently coloured. The longitudinally striped pattern of 
the elytra is peculiar to this species. 

Xanthophoea carteei, n.sp. 

Narrow, elongate, depressed. Prothorax sinuate posteriorly, basal angles 
rectangular; elytra lightly narrowed to base, interstices 3, 5 and 7 bearing fixed 
setae. Testaceous, disc of prothorax with an infuscate plaga on each side, elytra 
3-vittate, central vitta on interstice 1 of each elytron, lateral vitta on interstices 
6 — 8, an oblique broken fascia of piceous spots on each elytron from posterior 
third of lateral vitta to apex of sutural vitta; under surface wholly testaceous. 

Head convex (1.26 mm. across eyes), obliquely narrowed behind eyes, setu- 
lose; joint 3 of antennae finely setulose. Prothorax broader than long (1.25 x 
1.45 mm.), depressed and explanate at sides, lightly convex on disc, sparsely 
covered with small setules arising from very small punctures; sides rounded on 
anterior two-thirds, sinuate posteriorly; basal angles acute. Elytra widest about 
apical fourth (3.8 x 2.2 mm.); striae deep, crenulate; interstices hardly convex, 
distinctly punctate, 3 bearing six, 5 four, and 7 three strong punctures with 
maeroehaetae. Under surface, including prosternum and metasternum, sparsely 
setulose. d", with one seta at each side of apex of abdomen. Length, 6.5; 
breadth, 2.2 mm. 

Hah. — N.S. Wales: Gosford (Carter), Wahroonga (Sloane). Type in Coll. 
Sloane. I first received this species from Mr. H. J. Carter; subsequently I found 
a specimen in July, hiding in the leaves of a grass-tree {Xanthorrhoea hastilis). 

A distinct species allied to X wilsoni SI., from which it differs by size larger; 
basal angles of prothorax more acute; elytra more coarsely punctate, pattern 
different, interstices 2 — 5 testaceous on anterior two-thirds, and each with a small 
piceous spot on apical third (the spot on 2 united with the apex of the sutural 
vitta, that on 5 united with external vitta, that on 3 placed obliquely between 
those on 2 and 5, that on 4 opposite that on 2). 

Xanthophoea parallela Chaudoir. 
(= X. ornata Sloane.) 
I now feel no doubt but that X. parallela Chaud., (1872), is the same species 
as the one I subsequently named X. ornata (1910). It probably has a wide 
range; I described it from specimens found hiding in the leaves of the grass- 
trees at Berowra, near Sydney; there are specimens in the Macleay Museum 
ticketed King George's Sound. 



38 STUDIES IN AUSTRALIAN ENTOMOLOGY^, XVill., 

Xanthophoea LONGicoLLis Macleay (1864). 
(== X. ferruginea Chaudoir, 1872 = Demetrius rufescens Macleay, 1887). 

There seems to me no doubt but that Gymindis longicollis Mael. is the' same 
thing as X. ferruginea Chaud. and Demetrius rufescens Macl. It ranges from 
the Hunter River at least as far north as Cooktown. It is always obtained by be- 
ing beaten from the foliage of shrubs. 

Xanthophoea trivittata_, n.sp. 

?. Elongate. Head convex, obliquely narrowed behind eyes, impunetate, an- 
tennae with three basal joints pubescent; prothorax subquadrate, sides emarginate 
posteriorly, basal angles rectangular, disc impunetate; elytra parallel, crenulate- 
striate, interstices punctulate, 3 unipunctate about apical third; apex of abdomen 
(2) 4-setose; tarsi setulose on upper surface. Testaceous, elytra paler than head 
and prothorax; disc of prothorax with a black plaga on each side; elytra with 
three black vittae, central vitta on interstices 1 and 2, extending to apical sixth, 
lateral vittae on interstices 6 — 8, extending to extremity of 7, apex of abdomen 
infuscate. 

Head stout (1.7 mm. across eyes), transversely impressed posteriorly; orbits 
weU developed and oblique behind eyes. Prothorax slightly broader than long 
(1.8 X 2 mm,), a little wider across base (1.65 mm.) than apex (1.5 mm.); 
apex truncate, anterior angles obtuse; sides lightly rounded, lightly sinuate pos- 
teriorly; base bisinuate, squarely truncate on each side, basal angles rather pro- 
minent; lateral margins wide, punctate; anterior marginal seta before middle at 
widest part, very near edge; median line strongly impressed. Elytra subparallel, 
a little narrowed from posterior third to base, much wider than prothorax (5.5 
X 3.2 mm.) ; apex arcuate-truncate, external angles rounded off; interstices lightly 
convex, puncturation stronger on interstices 6 — 8 than on others. Length, 8 — 9 . 5 ; 
breadth, 3.2 mm. 

Hah. — Queensland: Stradbroke Island (H. J. Carter). I have examined three 
specimens given to me by Mr. H. J. Carter; owing to the two smaller specimens 
having their elytra spread I have only been able to measure the width of the 
largest specimen. 

A distinct species allied to X. parallela Chaud., from which it differs greatly 
by smaller size, head more strongly narrowed and not black behind eyes; pro- 
thorax more rounded on anterior part of sides, more strongly sinuate posteriorly, 
basal angles much more sharply marked, lateral channel much deeper, lateral 
border more widely and strongly reflexed; sutural vitta of elytra not reaching 
apex. 

Genus Phloeocarabus. 

Phloeocarabus parviceps, n.sp. 

Elongate-oval, depressed, upper surface glabrous. Head small, strongly 
constricted to a narrow neck; prothorax narrow, subquadrate; elytra lightly 
narrowed to base, lightly striate. Brownish testaceous; elytra with a wide in- 
fuscate fascia on apical laalf, this infuscation extending rather faintly forward 
along suture to base and spreading over interstices 1 — 4 near base, apex and a 
narrow lateral margin (inr-lnding interstice 9) testaceous; liead, antennae, and 
tarsi reddish. 



Bif I. G. SLOANE. 39 

Head obcordate (1.23 mm. across eyes), depressed, very finely shagreened 
aud microscopically punctate, strongly roundly-obliquely constricted behind; eyes 
prominent, rounded continuously with the curve of the posterior part of orbits. 
Prothorax depressed, subquadrate (1.25 x 1.65 mm.), ampliate behind apex, 
subparallel on sides, broadest at anterior third, wider across base than apex, 
finely shagreened; disc very finely transversely striolate; lateral margins wide; 
lateral basal impressions wide, concave; sides very little narrowed posteriorly, 
subsinuate before base; median line distinct; base lightly arcuate in middle, lightly 
oblique on each side; angles marked, but obtuse at summit. Elytra subquadrate 
with angles rounded (4.2 x 2.85 mm.), widest about middle; sides lightly 
rounded; apex wide, truncate in a light curve; striae fine, shallow, crenulate; 
interstices depressed, shagreened, sparsely microscopically punctate, 3 bipunctate, 
the punctures small (one just before, the other just behind the dark fascia), punc- 
tures of 9 small, not numerous, interrupted in middle. Length, 7.7; breadth^ 
2.85 mm. 

Eab.—^N.S. Wales. Type in British Museum ticketed "Illawarra, G. E. 
Bryant, 2.x. 08." I have seen it from the Blue Mountains. 

This species has somewhat the appearance of a species of Xanthophoea. 
The head is unusually small, and the prothorax unusually narrow, comparing it 
with Ph. mastersi Macl., it differs conspicuously by these characters, amongst 
others. It seems to come near Diabaticus collaris Black., (unknown to me irS 
nature) but evidently is differently coloured, and I expect it will prove to have 
the head smaller, neck narrower, prothorax less narrowed to base and less sinuate 
on the sides. Owing to the upper surface being shagreened its colours are rather 
dull. 

Genus Agonochila. 

AGONOCHILA OVALISj n.sp. 

Oval, rather convex. Head and prothorax piceous brown; elytra black and 
fiavous, the flavous area covering most of the disc, irregular on outer margin, 
broken into along basal part of suture and on middle of each side by intrusions 
of the black circumferential area ; a narrow broken zig-zag fascia extending back- 
wards in a faintly marked course across disc of elytra between inner sides of 
lateral intrusions from black ai'ea, lateral channel flavous; legs and antennae 
testaceous. 

Head smooth, with a few microscopic punctures. Prothorax transverse, much 
wider across base than apex, ampliate behind the widely rounded anterior angles, 
sparsely covered with short fine hairs; disc ratEer convex; sides explanate, finely 
punctate, not angulate at position of lateral seta, lightly obliquely narrowed to 
base; basal angles subrectangular, obtuse at apex. Elytra lightly convex, oval, 
densely and finely setulose-punetate ; apex wide; striae not marked, but interstices 
perceptible on disc. Length, 4; breadth, 1.75 mm. 

Hah. — Queensland. Type in British Museum ticketed "Kuranda G. E. Bryant 
l.vii.09." 

It is not unlike a species found by Dr. E. Mjoberg on the Bellenden Ker 
Mountains, which I have named A. intricata, but differs by having the discal 
flavous area of the elytra less decidedly divided by a distinct broken black fascia; 
size larger; head and prothorax less strongly punctate, prothorax with anterior 
angles more widely rounded, lateral margins much wider. It may be noted that 
the flavous discal area almosi touches the lateral margin about apical fifth. 



40 



NEW TERMITES FROM CENTRAL AND SOUTH-EAST AUSTRALIA. 

By Gerald F. Hill. 

(Twenty-one Text-figures.) 

[Read 28th March, 1923.] 

One new species of Coptotermes and two new species of Eutermes are 
described in this paper. With these additions to the list of Australian species, 
the former genus now contains six species and the latter twenty-eight species and 
one variety. 

I am indebted to the Authorities of the South Australian Museum and to 
Messrs. W. W. Froggatt, F. E. Wilson and W. F. Hill for the privilege of 
examining the material herein discussed. 

The methods of recording measurements and colours are the same as those 
referred to in previous papers. The Text-figures were drawn by the writer with 
the aid of a camera lucida and are from specimens in the type series. 

Coptotermes sedulus, n.sp. (Text-figs, 1-7.) 
Imago. (Text-figs. 1-4.) 

Colour: Very dark brown, head and pronotum darkest, leg's and wing veins 
a little lighter than tergites of abdomen. Antennae brown, first two joints very 
dark. Wing membrane fuliginous. 

Head (Text-fig. 1) rather small, moderately setaceous, rounded on the sides 
and behind, front flattened; fontanelle very small, oval; ocelli large, oval, well 
separated (0.048) from the eyes; eyes large (0.272 x 0.304), jDrominent, finely 
faceted, lower margin separated from the lower (lateral) margin of the head 
by a space of 0.329; Post-clypeus short, one-fourth as long as wide, light brown 
like tarsi; anteclypeus large, hyaline, anterior margin slightly produced in the 
middle; labrum large, rounded on the sides and in front; antennae (Text-fig. 2) 
19-jointed, 1st joint large, a little swollen at apex, 2nd half as long as 1st and 
much narrower, 3rd very short, less than half as long as 2nd and much narrower, 
smallest of all, 4th and 5th globose, equal. 

Thorax: Pronotum (Text-fig. 3) nearly as wide as head, concave in front, 
rounded on the sides, slightly sinuate behind, moderately setaceous like head; 
posterior margin of meso- and metanotura slightly more rounded than pronotum. 

Legs moderately short and slender, claws and tibial spurs large and dark 
coloured, tarsi pale coloured, fourth tarsal long and slender, longer than remain- 
ing joints together. 



BY GERALD F. HILL. 



41 



Wings: Wing-stumps setaceous as in pronotum, distal margin convex; mem- 
brane clothed with setae and densely but minutely sculptured (Text-fig. 4) ; costal, 
apical and distal half of the posterior margin very setaceous; costa and radius 
dark and very distinct, the latter setaceous to the apex of the wing and some- 
times obscurely branched towards the former, or to the median vein, or to the 
apex; median vein very distinct at the base but soon becoming invisible, that of 
the forewing branching within the wing-stump, that of the hindwing beyond the 
suture; cubitus with a varjdng number of branches, forked or simple, only the 
proximal short branches distinctly visible. 

Abdomen cylindrical, bluntly rounded at the apex, tergites clothed as in 
pronotum; cerci short and moderately stout. 

Measurements : 

Length with wings 15.00 mm.; without wings 8.00. 

Head: from posterior margin to apex of labium, long 1.36; from posterior 





ZXmiB33 




Coptotermes sedulus, n.sp. 

Text-figs. 1-4. Imago. — 1. Head from above; 2. antenna, proximal segments; 3. 

pronotum; 4. wing-membrane (greatly magnified). 
Text-figs. 5-6. Soldier. — 5. Head and pronotum; 6. antenna, proximal segments. 
Text-fig. 7. Worker. — Antenna, proximal segments. 



margin to clypeo-frontal suture, long 0.94; at and including eyes, wide 1.18. 
Antennae (19-jointed), long 1.92. 
Pronotum, long 0.70; wide 1.03. 
Tibia iii., long 1.00. 

Wings: forewings, long 11.00; wide 3.15; hindwing^, long 10.50; wide 3.30. 
Abdomen, wide 1.55. 

Soldier. (Text-figs. 5-6.) 

Colour: Head and antennae ochraceous, mandibles castaneous, pronotum 
bordered with yellow, legs and abdomen yellowish-white, the latter, as well as 



42 NEW TERMITES PROM CENTRAL AND SOUTH-EAST AUSTRALIA^ 

ineso- and metanotum, with an indistinct brown pattern along the dorso-median 
line. 

Head (Text-fig. 5) widest midway between base and fontanelle, rounded be- 
hind and on the sides, sloping in from the posterior margin of the antennal 
fossae to the base of the mandibles, with scattered long reddish setae; fontanelle 
relatively large (0.175 diam.), with dark coloured dorsal margin; anteclypeus 
hyaline; labrum long and narrow, acuminate; antennae (Text-fig. 6) 16- jointed, 
the 1st long and moderately slender, 2nd about two-fifths as long as 1st and 
much narrower, 3rd shortest of all, narrowed at the base, 4th as long as 2nd, 
oval. 5th as long as 4th, more rounded, 6th much larger and more stalked than 
5th, 7th turbinate. 

Thorax: Pronotum (Text-fig. 5) small and much narrower than head, deeply 
emarginate in front, slightly sinuate behind, the margin yellow, the whole upper 
surface with scanty reddish setae as on head ; meson otum about as wide as prono- 
tum, metanotum slightly wider, clothing and posterior margin of each similar to 
pronotum. 

Legs rather stout, clothed with numerous long and short reddish setae. 

Abdomen elongate, bluntly rounded at the apex, the segments clothed similarly 
to the legs; cerci large. 

Measurements of: 

C. sedulus C. lacteus C. michaelsend. 

Total length 5.30 — 4.30* 

Head, with mandibles, long .... 2.02 2.50 1.30 
„ , posterior margin to labral 

suture, long 1.36 1.60 — 

„ , posterior margin to anterior 

margin of fontanelle, long 1.27 1.41 1.22 

„ , wide 1.12 1.18 0.96 

Thorax and abdomen, long .... 3.66 — — 

Antennae, long 1-64 — 1.3C 

Pronotum, long 0.47 0.47 — 

wide • 0-82 0.89 — 

Mesonotum, wide 0.84 — — 

Metanotum, wide 0.94 — — 

Tibia iii., long 1.03 1.12 0.84 

Abdomen, wide 1 . 17 — — 

Worker. (Text-fig. 7.) 

Colour: Head yellow with ferruginous spot at postero-lateral angles of 
clypeus, remainder of insect almost uniform pale straw; head, thorax, abdomen 
and legs clothed with long and short, conspicuous, reddish setae. 

Head large, basal half from posterior margin of antennal fossae almost 
hemispherical, the widest part in line with the insertion of the antennae; post- 
clypeus large, anterior and posterior borders arcuate; labrum strongly convex, 
large, rounded on the sides and in front; antennae (Text-fig. 7) 16-jointed, 1st 
.joint large, narrowed in the middle, 2nd half as long as 1st, narrowed at the 
base, 3rd nearly as long as 2nd, showing faint indication of division at the 
proximal third, 4th and 5th very little longer than 3rd. 

Thorax: Pronotum much narrower than head, similar to that of soldier but 
slightly more rounded behind; mesonotum (0.94 wide) a little wider than prono- 
tum and a little narrower than metanotum (0.99 wide). 

* F. Silvestri. Die fauna Sudwest Australiens, Isoptera, Bd. 2, No. 17, 1909. 



BY GERALD F. HILL. 43 



Legs moderately long and stout, markedly setaceous; claws long and slender. 

Abdomen large, widest in the middle, tapered to the bluntly pointed apex; 
eerci large, narrow at the base, setaceous. 

Measurements : 

Total length, 6.25 mm. 

Head: base to apex of labrum, long 1.45; wide 1.14. 

Thorax and abdomen, long 4.80. 

Pronotum, long 0.47; wide .0.80. 

Tibia iii., long 1.00. 

Abdomen, wide 1.55. 

Described from a small nest series comprising workers, soldiers and two 
alate images (3.9.21), eight imagos captured on the wing, and numerous other 
series comprising workers and soldiers only. From the collectors' notes it is 
evident that this species closely resembles G. acinaeiformis Froggatt and C. lacteus 
Froggatt in its habits. It is probable that it is the species generally responsible 
for the destruction of fruit trees, vines, root crops, building timber, fence posts, 
etc., in Victoria, where it appears to be the predominant species of this genus. 
My collection of thirty nest series of Coptotermes from various parts of Victoria 
comprises twenty -nine series which I refer to the proposed new species and one 
series, from Rochester District, which may be referable to C. lacteus. 

Through the courtesy of Mr. W. W. Froggatt, I have been able to compare 
C. sedulus wnth the types of C. lacteus Froggatt and C. acinaeiformis Froggatt, 
from Shoalhaven, N.S.W., and Hall's Creek, N.W.A., respectively. In lacteus 
the head is hazel (Ridgway) ; it differs from sedulus otherwise in having a longer 
and broader head, larger pronotum, noticeably wider mesonotum and considerably 
longer (russet coloured) wings. C. acinaeiformis is a markedly distinct species. 
Professor Silvestri has kindly examined the alate form of sedulus and has ex- 
pressed the opinion that it is specifically distinct from both lacteus and michael- 
seni. The latter is the smallest of the Australian species and has two antennal 
joints less than sedulus in both imago and soldier. The heads of the soldiers of 
sedulus are markedly longer and more rounded than those of michaelserd, but 
shorter and narrower than those of lacteus. C. raffrayi Wasmann is a much 
larger and more robust species. The only remaining described Australian species 
is C. australis Walker, an inadequately described but apparently very distinct 
species from South Australia. 

Types in the writer's collection; paratypes in the collections of Professors 
F. Silvestri and S. F. Light. Locality. — Victoria: Belgrave, Healesville, Lakes 
Entrance (F. E. Wilson), Femtree Gully (F. P. Spry), Seaford, RiddeU (W. F. 
Hill). 

EuTERMBS FiELDi^ n.sp. (Text-figs. 8-13.) 

Imago. (Text-figs. 8-10.) 

Colour: Head and thorax dark castaneous; clypeus, antennae, legs and ter- 
gites of abdomen brussels-brown, stemites a little lighter; wings light smoky, with 
dark costa and radius. 

Head (Text-fig. 8) small, with scanty small pale setae (? abraded), broadly 
rounded behind, rather longer than wide, front concave; fontanelle small, in- 
distinct, in line with middle of eyes; eyes moderately large (0.282) and pro- 
jecting, situated near (.094) lower margin of head; ocelli small, narrow, widely 
separated from eyes; postclypeus very large and convex, more than half as long 



M 



NEW TERMITES FROM CENTRAL AND SOUTH-EAST AUSTRALIA, 



as wide (0.330 x 0.565), arcuate behind, nearly straight in front; anteelypeus 
very short; labrum short and broad, hardly covering apex of mandibles; an- 
tennae (Text-fig. 9) of doubtful number of joints, 1st large, broad; 2nd half as 
long as 1st, narrowed at base; 3rd short, narrowest of all; 4th and 5th about 
as long as 3rd but wider; 6th longer and wider than 4th and 5th; 7th and 8th 
larger than 6th, equal; other joints missing. 

Thorax (Text-fig. 8) : Pronotum large, a little wider than head, anterior 
margin slightly raised, sides sloping back sharply to slightly emarginate posterior 
margin; meso- and metathorax with rounded sides and broadly emarginate pos- 
terior margin. 

Wings : Wing-stumps small, equal, moderately setaceous, suture straight. 
Wings small; eosta and radius dark, media indistinct except at base; cubitus 
with about 12 inferior branches, only the first four or five of which are distinct. 





\ 



1 1 



-fc t 



10 




Eutermes fie''di, n.sp. 

Text-figs. 8-10. Imago. — 8. Head, pronotum, and posterior margin of meso- and 
metanotum; 9. antenna, proximal segments; 10. wing- 
membrane (greatly magnified). 

Text-figs, 11-13. Soldier. — 11. Head in profile; 12. head from above; 13. antenna, 
proximal segments. 



Wing-membrane (Text-fig. 10) clothed with numerous small setae; sculpturing 
apparently in the form of minute blunt spines (visible under high magnification 
only). 

Measurements : 

Length : with wings 13.50 ; without wings 8.00 mm. 

Head: from po.'^terior margin to apex of labrum, long 1.70; at and including 
eyes, wide 1.22. 

Pronotum: long 0.75; wide 1.24. 

Wings: forewings, long 9.00, wide 2.35; liindwings, long 8.75, wide 2. GO. 

Tibia iii., long 1.50. 

Abdomen, wide 2.25. 



BY GERALD F. HILL. 45 

Soldier. ( Text-figs. 11-13. ) 

Colour: Head uniform orange-rufous, thorax yellow-ochre, anterior half of 
pronotum darker than I'emainder, legs paler. 

Head (Text-figs. 11 and 12) small, rounded behind in dorsal aspect, snout 
short, about half as long as remainder of head; antennae (Text-fig. 13) of doubt- 
ful number of joints, 1st long and moderately stout, 2nd short and narrow, half 
as long as 1st and very much narrower, 3rd very long and narrow, nearly as long 
as 1st and about as wide as 2nd, 4th a little longer and wider than 2nd, more 
rounded, 5th and 6th longer than 4th, subequal. 

Pronotum small, a little more than half as wide as head, anterior half sharply 
bent up and narrowed anteriorly, posterior margin rounded, with hardly ap- 
preciable emargination ; mesonotuni much narrower than pronotum or metanotum. 

Legs moderately long, slender. 

Measurements : 

Head: long 1.27-1.30 mm.; wide 0.75. 

Pronotum: long 0.19; wide 0.47. 

Mesonotum, wide 0.33. 

Metanotum, wide 0.47. 

Tibia i., long 0.66. 

Affinities. — The imago of E. fieldi, n.sp. is very similar to that of E. hastilis 





' ^ 


■ ^-^ 


:r 


^ 








■^ 


-y 






-% 




r 


A 


^ y • 




4- 


-^ 


^ 

4 




^ ^ 
"f 


>- 
Y 


i 


> 






^ ^ 


^ 


f 






r 


If- 


-k 




14 


>< 


V 


1. "^ 


V 


%- 








16 

Euterme.3 tumuli Froggatt. 
Text-fig. 14. Imago. — Wing-membrane (greatly magnified). 
Text-figs. 15-16. Soldier. — 15. Head in profile; 16. antenna, proximal segments. 

Froggatt (identified for me by Mr. Froggatt from soldiers and workers only, 
collected at Stapleton, Northern Territory). It differs in having (1) the whole 
surface of the wing-membrane clothed rather densely with small setae (in hastilis 
~the basal half of the wing is without setae excepting for a few on each of the 
veins ; on the apical half the setae are more numerous, but not nearly so numerous 
as in fielS) ; (2) slightly smaller and less prominent eyes; (3) ocelli smaller and 
more widely separated from the eyes; (4) larger postelypeus; (5) lighter coloured 
mouth-parts; (6) darker sternum, and (7) larger pronotum. The soldiers of 
these two species differ more markedly, hastilis having paler head and very 
slender snout. The imago of fieldi also resembles E. tumuli Froggatt (types 
compared). The wings of the latter are slightly larger and darker and the veins 



46 



NEW TERMITES FROM CENTRAL AND SOUTH-EAST AUSTRALIA, 



are distinct throughout their length, the eyes are larger and the wing sculpturing 
entirely different (compare Text-figs. 10 and 14). The soldiers of fieldi and 
tumuli are easily differentiated by the antennae and the shape of the heads (com- 
pare Text-figs. 11 and 13 with 15 and 16). 

Described from a series, comprising 4 alate imagos, several nymphs and 
eight soldiers in alcohol, from the South Australian Museum collection. The 
specimens are in poor condition but are sufficiently perfect to be adequately 
described. 

Types in the South Australian Museum; paratypes in the author's collection. 

Locality.— Centval Australia: Tennant Creek (J. F. Field, May, 1907). 

EuTERMEs TRiBULis, n.sp. (Text-flgs. 17-21.) 
Queen. (Text-fig. 17.) 

Colour: Head, thorax, wing-stumps and abdominal tergites auburn; clypeus 




Eutermes trihulus, n.sp. 
Text-fig. 17. Imago. — Head, pronotum, and posterior margin of meso- andmetanotum. 
Text-figs. 18-20. Soldier. — 18. Head and pronotum in profile; 19. head from above; 

20. antenna, ten proximal segments. 
Text-fig. 21. Worker. — Head from above. 



amber-brown; anteclypeus, mouth-parts, under surface and legs lighter. 

Head (Text-fig. 17) moderately large, setaceous, hemispherical behind the 
eyes, clypeo-frontal .suture markedly concave, fontanelle very large, elongate-oval. 



BY GERALD F. HILL. 47 

about equidistant between clj^eo-f rental suture and posterior margin of head; 
eyes small (0.235 diam.)^ prominent, separated from the lower margin of the 
head by a space of 0.140; ocelli large, oblique, very widely separated from the 
eyes; elypeus moderately large, four-sevenths as long as wide, clypeal suture 
straight, anteclypeus very short, slightly arcuate in front; labrum about as long 
as elypeus, hardly covering apices of mandibles; antennae (mutilated) with 2nd 
joint small, less than half as long as 1st, and much narrower, 3rd very small, 
smallest of all, 4th and 5th rounded, equal. 

Thorax (Text-fig. 17) : Pronotum nearly as wide as head and similarly 
clothed, anterior margin very slightly sinuate, sides rounded to the emarginate 
posterior border; meso- and metanotum with posterior margin sinuate (as shown 
in Text-fig. 17) ; wing-stumps of the mesonotum larger than those of the metano- 
tum and not reaching the latter, surface setaceous, veins dark, suture straight. 

Legs short and moderately stout, setaceous. 

Abdomen : Tergites short. 

Measurements : 

Total length (about) 17.00 mm. 

Head: from posterior margin to apex of labrum, long 1.50; from base to 
clypeo-frontal suture, long 0.90; at and including eyes, wide 1.31. 

Pronotum : long 0.658, wide 1.034. 

Tibia i., 0.94. 

Soldier. (Text-figs. 18-20.) 

Colour: Head and antennae uniform amber-brown, anterior margin of prono- 
tum a little darker; thorax and legs buckthorn-brown. Head, thorax and abdomen 
scantily clothed with long and moderately stout setae. 

Head (Text-figs. 18 and 19) long and slender, snout nearly as long as re- 
mainder of head, head slightly constricted on the sides and above; antennae 
(Text-fig. 20) 14- jointed, long and slender, 1st joint long and rather stout, 2nd 
about half as long as 1st and very narrow, 3rd longer than 2nd and about as 
wide, 4th as long as 1st, slender, 5th and 6th equal, a little shorter than 4th, 7th 
as long as or longer than 1st and 4th, remaining joints a little shorter than 7th. 

Thorax: Pronotum very small, saddle-shaped, anterior half sharply bent up 
and darker in colour than posterior half, anterior margin fringed with long setae. 

Legs long and slender, clothed scantily with long setae; claws very long and 
slender. 

Abdomen clothed with markedly long and stout setae; c.erci very long and 
slender. 

Measurements : 

Head: long 1.645 mm.; wide 0.730. 

Pronotum: long 0.188; wide 0.470-0.517. 

Tibia iii., long 1.270. 

Worker. (Text-fig. 21.) 

Colour: Head mummy-brown; elypeus and mouth parts buckthorn-brown. 

Head (Text-fig. 21) large, broadly rounded behind, widest at the insertion 
of the mandibles, frontal and transverse sutures very distinct, clypeo-frontal 
suture markedly concave; elypeus small, not strongly convex, clypeal suture 
straight, anteclypeus small, straight in front; labrum small, rounded, not covering 
apices of the mandibles. 



48 STUDIES IN AUSTRALIAN ENTOMOLOGY,, xviii. 

Thorax: Pronotum very small, saddle-shaped, anterior half bent up sharply, 
rounded in front, with deep emargination in middle, the whole surface with 
scattered stout setae. 

Measurements : 

Head: base to apex of mandibles, long 1.645 mm.; base to clypeo-frontal 
suture, long 1.222; wide 1.550. 

Tibia iii., long 1.410. 

The specimens at my disposal are in poor condition, but as all the castes are 
represented and the species is a very distinct one I have not hesitated to describe 
it as new. 

Described from a series comprising one true queen, one neoteinic queen, 
about twelve soldiers and one worker, from the South Australian Museum 
collection. 

Types in the South Australian Museum; paratypes in the writer's collection. 

Locality. — Central Australia: Tennant Creek (J. F. Field) 



49 



STUDIES IN LIFE-HISTORIES OF AUSTRALIAN DIPTERA 
BRACHYCERA. 

Part i. Stratiomyiidae. 

No. 4. The respiratory systera in larva, pupa and imago of Metoponia ruhriceps 

Maequart. 

By Vera Irwi^^-Smith, B.Sc, F.L.S., Linnean Maeleay Fellow of the Society in 

Zoology. 

(Fifty Text-figures.) 

[Read 18th April, 1923.] 

Page. 

Introdviction; Historical review; Methods of Study 49-52 

The tracheal system of the larva 52 

• (a. Tracheae, 52; b. Air chamber, 53; c. Posterior spiracles. 55; 
d. Anterior spiracles, 55; e. Head spiracles, 57; f. Lateral 
spiracles, 57) 

Development of secondary spiracular papillae .57 

Pupal Metamorphosis 60 

The tracheal system of the pronymph 61 

The tracheal system of the nymph 65 

Development of the secondary ^abdominal spiracles 66 

The tracheal system of the proimago 69 

The tracheal system of the imago 71 

(a. Anterior thoracic spiracles, 71; b. Posterior thoracic spiracles, 
71; c. Abdominal spiracles, 72; d. Tracheae, 73) 

Summary 75 

Note on the comparative morphology of the spiracular apparatus . . 77 

Introduction. 

The respiratory system of this species Avas dealt Avith very briefly in the 
general account of the life-history given in the first paper of the series (Irwin- 
Smith, 1920). The further investigation of certain structures then referred to 
has led to a detailed study of the whole system in larva, pupa and imago, and 
has resulted in the discovery of features of great interest and significance, which 
axe probably common to the whole Stratiomyid family, although they have 
hitherto escaped the attention of Dipterologists. The present paper gives the 
results of the investigation, and forms a contribution to the important subject 
of the post-embryonic development and comparative morphology of the re- 



50 STUDIES m LIFE-HISTORIES OF AUSTRALIAN DIPTERA BRACHYCERA, 

spiratory system in Diptera and in insects in general, upon which it is hop^d 
that some additional light will be thrown by the new information now pre- 
sented. 

Historical review. 

Haliday, as long ago as 1857, referred to the ''complexity of the respiratory 
apparatus, superadding the series of lateral spiracles to the curious fringed bell 
for air which usually surrounds the posterior opening" as one of the characters 
which serve to make the larvae of the Stratiomyiidae "about the most perfectly 
organised among the Diptera." But, in spite of this, all workers on the group 
up to the present time have confined their attention to the external features 
of the system, and existing descriptions consist merely of an enumeration of the 
spiracles or of a very brief account of their appearance and arrangement. No 
attempt has been made to examine the internal structure, or to follow out its 
development in pupa and imago. 

Most v.-riters follow Brauer (1883) in describing the larvae as either amphi- 
pneustic, or peripneustic with stigmata on 1st and 3rd to 7th or 4th to 7th seg- 
ments behind the head. Malloeh's diagnosis (1917) of the larval characters in- 
cludes the statement "lateral metathoracic and abdominal spiracles present or 
absent." From the study of the larva of Metoponia ruhriceps it will be seen that 
this question of t'he presence or absence of lateral spiracles is one of consider- 
able importance. But on this point descriptions of the same genera given by 
different authors frequently fail to agree. Some genera, e.g., Pachygaster and 
Xylomyia, are generally described as amphipneustic ; in other genera, e.g., 
Oxycera, the metathoracic and seventh, abdominal spiracles are said to be miss- 
ing. But, while some observers have failed also to detect lateral spiracles in 
Chloromyia, Michrochrysa and Sargus^ Lundbeck (1907) says that extremely 
small spiracles are present on the first seven abdominal segments in these genera, 
though metathoracic spiracles are missing; and, from accounts given by Lund- 
beck and many other workers, it seems certain that the genera Stratiomyia and 
Odontomyia possess lateral spiracles, although Malloch says they cannot be dis- 
tinguished. Both Malloch and Lundbeck, however, are of the opinion that even 
when lateral spiracles are undoubtedly present, they are probably not functional, 
ajid Lundbeck thinks that the prothoracic spiracles also are not in function, con- 
cluding "thus v.'ith regard to the tracheal system the larva could be termed meta- 
pneustic." 

Now, a careful examination of the larva of Metoponia ruhriceps has shown, 
that the lateral spiracles in this species are not only functional, but are of great 
importance in connection with the subsequent development of pupal spiracles. 
Brauer, in 1883, had already stated that the Stratiomyid nymph is always peri- 
pneustic, and that the first to the sixth abdominal stigmata are bound with the 
corresponding stigmata of the larval skin through tracheae. If, as is most pro- 
bable, the pupal spiracles are developed in all the genera in the way described 
liere for Metoponia, it seems safe to conclude that the Stratiomyid larva, too, 
must be always peripneustic, and that, where larvae have been described as amphi- 
pneustic, tlie lateral spiracles have been overlooked. This view is supported by 
Lundbeck's statement that "the spiracles are often so indistinct and especially 
just on metatliorax and towards the apex of abdomen, that it may be very 
difficult to decide their actual number." 

Meijere, in his elaborate account of the stigmata of Dipterous pupae (1902), 
mentions the Brachyeera very briefly, and devotes only a few lines to the 



BY VERA IRWIX-SMITH. 51 

Stratiomyiidae. He refers to Brauer's statement, and expresses the decided 
opinion that if, as Brauer affirms, the abdominal stigmata of larva and pupa are 
bound together by tracheae, the pupa is not, on this account, really peripneustic, 
since the stigma scars of the pupa are closed. "Dass durch dieselben noch 
einiger Luftwechsel stattfindet, giaube ieh jedoch nicht, weil die zusammenge- 
fallenen Stigmennarben dieses wohl uberhaupt nicht zulassen" (p. 640). 

My investigations have proved that Meijere was wrong in this. From the 
description of them given in this paper, it will be seen that the pupal spiracles 
are highly developed structures, and are certainly open and functional. 

With the exception of Brauer and Meijere no one has paid any attention to 
them. But several authors have noted papilla-like markings on Strathiomyid 
larvae without recognising their significance. Malloch (1917) describes and 
figures ''a small round wart just posterior to each spiracle" on abdominal seg- 
ments 2-5 of Hermetia illucens, and, in his figure of Stratiomyid larva, genus 
incertus 1, shows very distinct papilla-like folds in the same position, though he 
does not mention them in the text. Hart (1895) mentions "a dark dot surround- 
ing the spiracles" on the larva of Stratiomyia norma; Austen, in his account of 
Xylomyia maculata (1899), saj^s "Apparently the larva is amphipneustic .... 
On each side, however, of the first six abdominal segments, immediately behind 
the tumid lateral ridge, and so in the angle which each segment forms with the 
next, I observe a small papilla. On examining the larva skin vv^ith a microscope 
I cannot detect an aperture in any of these papillae, but it seems in the highest 
degree probable that they represent stigmata v\'hieh have disappeared, but were 
functional in the larvae of ancestors of the existing species of Xylomyia." Lund- 
beck, with reference to the same genus, observes (1907) "The larva is generally 
considered amphipneustic, as it has posterior and terminal spiracles, but I de- 
tected on the first seven abdominal segments very small and indistinct spiraeular 
plates, lying on each side of the segments, somewhat inwardly to the base of the 
lateral bristle; these spiracles are certainly not in function. The small papillae 
mentioned by Austen certainly have nothing to do with spiracles; such papilla- 
like folds are found in several larvae of Stratiomyids." 

The papillae which these authors saw are undoubtedly of the same nature 
as papillae which have been found in the same position on the larva of Meto- 
ponia. Their structure and development are described in detail in the present 
paper. It is sufficient to state here that, while they are not, as Austen thought, 
vestigial spiracles, they are, contrary to Lundbeck's opinion, directly connected 
with the respiratory system. It will be shown that each papilla marks the posi- 
tion of a pupal spiracle, and forms a protective covering for it during its 
development in the larva. 

The spiracles of the pupa and all the other parts of the respiratory system 
are, so far as I am aware, described hei'e for the first time. 

Methods of study. 

The study of the tracheal system in all stages has been carried out almost 
entirely by means of dissections, and the examination of whole specimens or of 
portions of them cleared and mounted in Canada balsam. The cuticle, especially 
of the larva, is so difficult to cut that sections are usually badly fractured, and 
even when successfully made, they are only useful for the study of histological 
details of spiraeular structure. The tracheal vessels are not easily traced in them, 
and for the working out of the entire system, cleared preparations of the whole 
insect have proved most useful. 



52 



STUDIES IN LIFE-HISTORIES OF AUSTRALIAN DIPTERA BRACHYCERA, 



Speeimens in every available condition have been examined, moult larva 
skins and empty pupa eases, preparations of artificial exuvia by treatment with 
caustic potash, and larvae, pupae and imagines in either salt solution, alcohol, 
glycerine, cedar oil, clove oil or canada balsam. Most of the larvae used were 
fixed in hot Carl's solution (absolute alcohol 15 parts, concentrated formol 6 
parts, glacial acetic acid 2 parts, distilled water 30 parts), and transferred to 
70% alcohol after twenty-four hours, the integument being slit or pricked to 
allow the fluid to penetrate; the pupae were removed from their pupa eases 
before being fixed; the imagines, collected in killing bottles, were stored in 70% 
alcohol, or chloroformed and dissected at once in salt solution. 

Useful observations were made from some of the specimens treated from 
three to six days in a 10% aqueous solution of caustic potash, and finally mounted 
in Canada balsam; but the most uniformlj^ satisfactory results T,^ere obtained from 
specimens dehydrated in alcohol, cleared in, cedar oil (mixed with absolute alcohol 
in gTadually increasing strengths up to the pure oil), and mounted unstained in 
Canada balsam. 

Cedar oil proved a better clearing agent than clove oil, and it was found 
that at all stages of development the tracheal system, the major vessels at least, 
showed up better in this medium than in any other tried, though for the com- 
plete elucidation of some features it was necessary to examine different specimens 
treated by one or other of all the various methods tried. For the examination 
of the finer details, the portion of the body required was removed with fine 
scissors or needles under a binocular dissecting microscope, and, after the neces- 
sary treatment, mounted separately in canada balsam. 

All the text-figures were prepared by the writer from drawings made at 
stage level with the aid of a Zeiss- Abbe camera lueida and Zeiss and Reichart 
oculars and objectives. 

The Tracheal System of the Larva. 

Tracheae. — There are two main tracheal trunks which lie in a dorso-lateral 
position and extend through the whole length of the body from the anterior 
spiracles to the posterior air chamber, diminishing in calibre from the posterior 
to the anterior end. A short arched transverse commissure connects them dor- 
sally at the anterior border of each body segment from the second to the tenth 
(Text-fig. 48). The commissure for the first segment occupies a deeply ventral 
position (Text-fig. 1, v.c.) ; if there is also a dorsal commissure, it is not dis- 
tinguishable among the crowd of fine tracheal branches in this region. No cor- 
responding commissure could be detected in the eleventh segment. 

Just behind the commissure in each abdominal segment, a fairly stout branch 
arises from the outer side of eacli trunk, and runs forward to junction with the 
corresponding branch in the next segment (Text-fig. 4, s.l.). In this way a series 
of closed segmental loops is formed along the sides of the body, the last loop (in 
the eighth abdominal segment) being formed by a large branch given off just 
above the posterior spiracle (Text-figs. 5, 48, s.l.). In the third thoracic seg- 
ment similar branches are given off, but they are very indistinct, and could not 
be traced to their connections; in the second segment, lateral branches could not 
be detected. 

From the under side of each loop three main branches arise, wliich pass out- 
ward and downward to form, with their ramifications, the principal supply to the 
body wall and viscera. From the main trunks themselves very few branches are 
given off, except at the anterior and posterior ends of the body. It is these 



BY VERA IRWIX-SMITH. 



53 



ajDical regions which receive the most extensive supply of tracheae. At the an- 
terior end, just behind the ventral commissure, a stout trachea is given off from 
the inner side of each trunk, which dips ventrally, and immediately divides into 
a cluster of radiating branches, spreading out fan-wise, and dividing into 
tracheoles to supply all the region of the head, pharynx and masticatory ap- 
paratus. At the posterior end, large branches are given off from the main 
trunks internally and externally just above the posterior spiracles, the principal 
external branches being formed in association with the segmental loops (Text-fig. 
5). The general trend of all these branches is upwards, away from the air 



as- 







Text-figures 1-5. Larval respiratory system. 

1. Spiracles and tracheae of prothorax, ventral view. (x 40). a.s., anterior 
spiracle and stigmatic chamber, seen from the back; v.c, ventral commissure. 
2. Anterior spiracle, front view, (x 160). 3. Back view of the same, (x 160). 
e., stigma excavation; s.c, stigmatic chamber. 4. Tracheal system of abdominal 
segment, lateral view, (x 25). d.c, dorsal commissure; l.s., lateral spiracle; s.l., 
lateral segmental loop; t., longitudinal tracheal trunk. 5. Termination of tracheal 
system at posterior end, dorsal view, (x 40). a.c, air chamber; h., sensory hairs; 
S.I., segmental loop. 



chamber, but their finer ramifications extend in all directions through this region 
of the body. 

The air chamber is situated dorsally behind the level of the anus (Text-fig. 
6). It opens by a narrow transverse slit in a deep groove which is slightly pre- 
apical, and, as explained in a previous paper (Irwin-Smith, 1920) is probably 
the dorsal incisure between the fused eleventh and twelfth body segments. In 
mature larvae, the aperture is .25 to .32 mm. wide. The area surrounding it is 



54 



STUDIES IN LIFE-HISTORIES OF AUSTRALIAN DIPTERA BRACHYCERA, 



heavily chitinised, and on each lip there is a row of very fine, short, colourless 
hairs, which together form a slight fringe over the opening, not sufficiently long 
or dense to screen it effectively (Text-figs. 7, 8). The eight larger hairs, or 
bristles, found in association with the posterior groove are figured in Text- 
fig. 5, h. They are probably of the nature of sensory hairs. 

When the body is split open by a longitudinal section through the groove 
and aperture, the air-chamber is exposed intact, on the dorsal portion. Its walls 
are thin and transparent, and only show up well when stained. In mature 
larvae it is of relatively considerable extent, .32 to .40 mm. long, and .25 to .32 







Text-figures 6-9. Larval respiratory system. 

6. Lateral view of posterior end, showing aperture of air chamber, (x 40). 
7. Aperture, (x 133). 8. The same, viewed from posterior end. (x 133). 9. 
Longitudinal section through air chamber, dorsal side, showing muscles attached, 
(x 63). 



mm. in maximum width, and, by the action of numerous muscles attached to its 
walls, it appears to control the admission of air to the tracheal trunks. The re- 
spiratory muscles are arranged in groups, which stretch in all directions between 
the chamber and the body wall; some are shown in Text-fig. 9. One power- 
ful group extends from the anterior end of the chamber to the dorsal wall; one 
group is inserted on the dorsal side and another on the ventral side of each 
spiracular opening, extending out laterally; two muscles just below the spiracle 
are attached vcntro-laterally to the body-wall; while other muscles connect the 



BY VERA IRWIX-SMITH. 55 

dorsal and ventral walls medianly with the corresponding body-walls. The 
points of attachment of the muscles are seen on the walls of the chamber as 
groups of oval brown patches. The width of the slit-like aperture of the chamber 
is evidently controlled also by the action of muscles. 

The posterior spiracles, which open anteriorly into the air chamber, have all 
the appearance of spiracles opening on an external surface of the body (Text- 
fig. 10). The large open mouth of each is circular, .10 to .13 mm. in diameter, 
with a thick, prominent rim, having a beaded ornamentation on its upper sur- 
face. The number of "beads" varies from 40 to 50. The first portion of the 
tracheal trunk, to a distance of .06 to .11 mm. from the rim, is modified to 
form a sieve-chamber, being provided with a number of short, stiff, rod-like 
setae, which project inwards from the cuticular Hning of the wall. The rods do 
not extend right across the lumen of the tube, but, in the sections examined,, the 
passage betn^een them is found to be occluded by a lump of granular material, 
in which the ends of the rods are embedded (Text-fig. 11). This granular matter 
has the appearance of a foreign substance, rather than an organic part of the 
structure. 

The anterior spiracles are present in the newly hatched larva, and retain the 
same form and structure throughout the whole larval life. They are many times 
larger than the lateral spiracles, and form conspicuous objects on the dorso- 
lateral margins of the prothorax. In mature larvae each measures from .100 to 
.150 mm. in length, and from .085 to .100 mm. in maximum width, and con- 
sists of two distinct regions (Text-fig. 2). The upper part, which protrudes 
above the surface of the integument, is disc-shaped, from .070 to .080 mm, wide, 
and from .040 to .055 mm. long, and is formed of very dense, dark brown 
chitin, with smooth upper face and even outline. It bears two large, oblique 
slits, ..033 to .044 mm. long, which converge posteriorly, and are separated by a 
narrow partition, incomplete at the base. Below this disc, there is a shield- 
shaped area of lighter brown chitin on a level with the body cuticle, with a 
roughened surface, having petal-like folds which converge on a narrow excava- 
tion towards its middle. The shield serves to protect the stigmatic chamber into 
which the two oblique slits open, since it covers exactly the area occupied by the 
latter. 

The stigmatic chamber forms the terminal portion of the main tracheal 
trunk, and corresponds to the "felted-chamber^' (P^ilzkammier) descaibed by 
Meijere (1895), It is certainly functional, but in it the tracheal walls have 
been modified to form a complicated occluding apparatus, evidently for the ex- 
clusion of foreign matter. It is a compound, cylindrical chamber, .080 to .092 
mm. long, and .040 to .050 mm. at its greatest diameter, where it junctions with 
the trachea proper (Text-fig. 3). It is filled with a mass of clear, delicate, chitin 
rods, embedded in an extremely fine meshwork, of which the structure is not 
clear, even under high magnifications. The rods are attached to the chitinous 
intima, and extend transversely and slightly upwards across the chamber, con- 
verging on an indentation along the upper wall, where it is slightly thickened, 
and covered by a mass of granular tissue (Text-fig. 13). In this a fine lumen 
may be detected. From the base of the chamber, in the region of this inden- 
tation, a narrow, simple tube is given off, which traverses the granular tissue, and 
is attached to the excavation on the chitinous shield, which constitutes the so- 
called "stigma scar" (Stigmenarbe) of Meijere. 

This tube has been already observed in a Stratiomyid larva, Pachygaster 
minutissima, by Tragardh (1914), and in many other dipterous larvae. Meijere 



56 



STUDIES IN LIFE-HISTORIES OF AUSTRALIAN" DIPTERA BRACHYCERA_, 



coasiders that it is the original tracheal tube, round which the stigmatic chamber 
has been formed as a secondary growth, a conclusion which seems to be borne 
out by a study of the lateral spiracles of the Metoponia pupa. It will be seen, 
from the description given later, that these are very similar structures, and they 
are certainly secondary formations growing up round the primary lateral spiracles. 
But if this is the case with the prothoracic stigma of the larva, the original 
tracheal ending must be a purely embryonic structure, since the compound cham- 
ber is already present when the larva emerges from the egg. On the other hand, 
there is some evidence that the pronymphal opening of the anterior spiracles is 
formed in the position of the larval stigma scar, so that the scar may be rudi- 
mentary rather than a vestigial structure. 







Text-figures 10-15. Larval respiratory system. 



10. Posterior spiracle, (x 133). 11. Longitudinal section through posterior 
spiracle, (x 2.53). 12. Lateral (abdominal) spiracle, (x 387). 13. Transverse 
section through stigmatic chamber of anterior spiracle, (x 387) . 14. Transverse 
section through lateral body wall in region of stigmatic chamber, (x 167). l.rn., 
lateral muscle; s., stigmatic shield. 15. Transverse section through the same, m 
region of junction of stigmatic chamber and trachea, (x 167). r.m., respiratory 
muscle . 

The stigmatic chamber is enclosed in a very thick layer of hypodermis, com- 
posed of a single row of long, narrow cells with large nuclei, which at the sides 
become continuous with the hypodermis of the body wall. 

As with the posterior air-chamber, the admission of air to the prothoracic 
stigmata is evidently controlled by the action of respiratory muscles. Muscles 
are atta/^hed to the junction between the trachea and stigmatic chamber, and ex- 



BY VERA lEWIiv'-SMITH. 57 

tend between it and the ventral body wall (Text-fig. 15). In addition, there are 
transverse muscular bands extending across the body in the lateral area occupied 
by the stigmatic chamber, which partially shut it off from the general body 
cavity, and, by their contractions, evidently vary its size and shape (Text-fig. 
14). 

Head spiracles. — The occurrence of spiracles on the head of the Stratiomyid 
larva is not mentioned by any previous worker. But a pair of structures which 
have all the appearance of spiracles is certainly present in this species. They 
are situated ventrally at the base of the lateral "bosses," and are referred to in 
the first three papers of this series (Irwin-Smith, 1920, 1921). They are figured 
in Part i. (PI. xxvii.), Part ii. (Text-fig. 4), and Part iii. (Text-fig. 13-17). 

Lateral spiracles. — Spiracles are present on the metathoracic and first seven 
abdominal segments. They are extremely small, and, under low magnification, 
can only be detected as a small dark spot on the border of each lateral gToove, 
towards its anterior margin. On the mesothorax, on which they do not occur, 
there is, in a con'esponding position, a small colourless hair. When highly mag- 
nified, the spiracles are seen to be cylindrical, chitinous structures, which pro- 
ject slightly (.01 mm.) above the surface of the integument, and have a diameter 
of .018 to .02 mm. (Text-fig. 12). Each has a distinct cavity, and is connected 
with the central tracheal system by a fine tube, which arises from the fork junc- 
tion of the segmental loops. In this tube a distinct lumen, which appears to be 
continuous throughout its length, is visible. As far as could be ascertained by 
careful microscopic examination, without the aid of serial sections, there is no 
"plugging" of the spiracle, nor solidification of any portion of its air tube, such 
as is described by Carpenter and Pollard (1918) for the Hypoderma laxva. 

It would appear, therefore, that the lateral spiracles are functional, contrary 
to the opinion of most workers on the Stratiomyiidae. But they are so minute 
that they cannot play a large part in respiration, and no valvular arrangement 
is discernible in connection with them. 

Development of secondary spiracular papillae. 

In the immediate vicinity of each abdominal spiracle, in the older larvae, a 
dark chitinous fold in the integument is plainly visible. As already stated, these 
structures have been observed in other Stratiomyid larvae, but their true nature 
has not been recognised hitherto. Close investigation has now proved that they 
are formed in direct relation with the formation of rudimentary pupal spiracles 
in the laxva, and that they undergo a process of continuous development through- 
out the whole of larval life. 

When large numbers of larvae are examined it is seen that there is a marked 
difference in this region in larvae of different ages; but the position of the 
secondary formation is clearly indicated on the larval cuticle from the earliest 
stages of growth. Immediately behind the primary (larval) spiracles, in the 
young larva, the calcareous hexagonal plates are smaller than elsewhere on the 
body, and at a distance of .Q5 to .07 mm. postero-dorsally from the spiracle, 
there is found a circular area of minute, delicate, rounded plates (Text-fig. 16). 
This region is found only on the segments which, in the pupa, bear the complex 
spiracles, i.e., the fii-st to the sixth abdominal. On the metathoracic and seventh 
abdominal segments, where no change occurs, the spiracular area presents the 
appearance shown in Text-fig. 21, in which large hexagonal plates of the regular 
character occur all round the spiracle. 

The next stage in the development of the secondary formation is shown in 



58 



STUDIES IN LIFE-HISTORIES OF AUSTRALIAN DIPTERA BRACHTCERA_, 



Text-fig. 17. It takes the form of a slight thickening of the plates of the 
circular area, and a folding over of the plates bordering it on the side remote 
from the spiracle, with a deposition of chitin along one side. In this way a 



cm 








18 ^^ 



50^^001 




Text-figures 16-22, shoAving lateral spiracles of larva, and development of 
secondary papillae (16-19). 

16. First stage. 17. Second stage. 18. Third, stage. 19. Fourth stage, on 
larva. 20. The same region on puparium, showing neck of pupal spiracle pro- 
truding through cuticle. 1., larval spiracle; p., pupal spiracle. Figs. 16-17 (x 
290), figs. 18-19 (x 305): fig. 20 (x 190). 21. Lateral spiracle of meta- 
thoracic segment, (x 290). 22. Dorso-lateral view of abdominal segments, show- 
ing position of secondary papillae at the fourth stage, (x 30). 



well-defined luie comes to be formed, beyond which the cuticle is slightly de- 
pressed. 



BY VERA IRWIN-SMITH. 59 

At a later stage (Text-fig. 18), the depression is accentuated into a definite 
trough, over which, on one side, the rim of chitinised plates projects. The cir- 
cular area has grown out into a rounded boss, on which the plates have become 
modified into a rosette-like arrangement of narrow, projecting leaves and spines, 
while on the rim of the trough immediately behind the boss, a cluster of larger 
spines takes the place of the ehitin plates. 

At a still later stage (Text-fig. 19) the whole area along the margin of the 
trough has become a deep brown, due to a heavy deposition of ehitin, the boss 
has become more prominent, and the spines stouter and stronger. The trough 
is .10 to .13 mm. long, and the secondary formation is now much more con- 
spicuous than the primary spiracle. 

Just below this region, in larvae which are still comparatively young, the 
pupal spiracles can be seen in process of formation. Their development will be 
described later, in connection Avith the pupal metamorphosis. The secondary 
papilla is evidently a protective covering for the spiracle during its formation, 
and the prominent boss marks the position of the neck of the spiracle. It does 
not perforate the cuticle during the larval stage, and no aperture could be de- 
tected in this region in any of the larval preparations examined, but at pupation 
it becomes visible on the external surface. When the puparium is removed from 
the pupa and prepared for microscopic examination, it is found that the neck 
of the spiracle is now protruding through the integument, and projects for a 
distance of .03 to .04 mm. over the trough (Text-fig. 20). The pupal spiracles, 
therefore, are already well developed and functional at the onset of pupation. 

Details of the secondary papillae are difficult to make out on the puparium. 
On the living pupa it is so dry and opaque, and usually so covered with grit, 
that little or nothing can be seen of it. A mass of dirt fills the trough, and 
covers the position occupied by the spiny plates in the lar\'a, and, even when the 
puparium is removed, and cleared for mounting, it is almost impossible to clean 
this away. But, as far as can be ascertained, the spiny formation disappears at 
the end of the larval period. 

The different stages observed in the development of the secondary papillae 
apparently represent successive larval instars, and, when more information is 
available, they should afford a good indication of the age of the larva. At pre- 
sent it is not possible to determine at what age the development begins, or the 
length of time the larva remains in each stage. Unfortunately, the papillae, 
especially in the earlier stages, are most difficult to see on the living larvae : they 
can only be detected in strong sunlight, and after long familiarity with their 
appearance in microscope preparations. 

Among the preserved specimens examined, most of those under 6 or 7 mm. 
long were still at the fij:st stage (Text-fig. 16) ; some larvae of 7.7 and 8.3 mm. 
lengths Avere at the second stage (Text-fig. 17) ; while others of 7.8 and 8.0 mm. 
were at the third stage (Text-fig. 18) ; others, again, of 8.3 mm. were at the 
fourth stage (Text-fig. 19), as were all those over a length of 8.5 mm. 

Seven larvae between 8 and 10 mm. long, which were collected in the field 
on November 4th, were already at the fourth stage, the shorter and more slender 
specimens being evidently males. They were kept alive, but went through the 
whole spring season without pupating, and are still active larvae (February 1st, 
1923). [See also postscript, jd. 81.] 

As it has been proved that the imagines only appear twice a year, in 



60 STUDIES IN LIFE-HISTOEIES OF AUSTRALIAN- DIPTERA BRACHYCERA^ 

November and April, it is evident that they cannot pupate now before April at 
the earliest, so that the fourth stage must occupy at least six months. 

Pupal Metamorphosis. 

Lowne (1892) recognises three distinct stages in the pupal development of 
the blowfly, to which he gives the names pronymph, nymph and proimago. He 
distinguishes as the pronymph the stage at which the head and thoracic ap- 
pendages are still invaginated, and the larval organs are undergoing histolysis, 
after the shedding of the larval skin to form the puparium and the enclosure of 
.the body in a temporary membrane, the paraderm. He considers that the nymph 
takes the place of the pronymph when the limbs and head are evaginated and the 
paraderm has been replaced by the pupal ectoderm, and applies the term pro- 
imago to the pupa from the period at which the pupal ectoderm is separated as 
the pupal sheath, to the escape from the puparium. He says "The changes which 
occur before the formation of the pupa sheath correspond nearly with those 
which take place in the caterpillar before the ecdysis of the last larval skin, 
whilst those which occur after its formation correspond with the changes which 
take place in the chrysalis, or nymph stage of the Lepidoptera." And he speaks 
of the tracheal system as it exists before "this period of virtual ecdysis," i.e., 
the separation of the pupa sheath, as the tracheal system of the njonph, and the 
new tracheal system which is subsequently formed, as the tracheal system of the 
proimago. 

Although the larval organs of the StratiomyUdae do not undergo the com- 
plete histolysis which is found in the Cyelorrhapha, the method of pupation in 
this family corresponds more closely with that of the Cyelorrhapha than does 
that of any other of the Brachycera. The same number of cuticular skins is 
found covering the ripe pupa, viz., pupa case, pupa sheath and epidermis of the 
imag'o, representing three distinct ecdyses, so that Lowne's three stages are clearly 
recognisable in the pupal period, and, for convenience, his terms can be applied 
to the different stages of tracheal development. 

The majority of the pupae of Metoponia rubriceps examined were already 
in the proimago stage; a few were obtained in the nymph stage, and one, only, 
at the pronymph stage. This last specimen I had the good fortune to find just 
after it had been attacked and killed by a fungal growth which had extended all 
over the tracheal system. The other tissues had already begun to decay, but the 
entire tracheal system, to its smaller ra,mifieations, was revealed in dark outline 
by the fungus with the clarity of an injected blood-vascular system. It is figured 
in Text-figs. 23, 27, 49. It is apparently at a late period in pronymphal de- 
velopment. The legs are already exposed, but the head is still invagiaated, and 
the cavity at the anterior end, within which it is withdrawn, contains the exu- 
viated pharynx and masticatory apparatus of the larva. The pupa sheath is not 
yet formed, and the body is still covered by an integument, the external layer of 
which consists of a thin, transparent cuticle. Whether this integument is the 
"paraderm" of Lowne, or the pupal ectoderm, there is not sufficient evidence to 
prove; the tissues are in too decayed a state for detailed histological examina- 
tion. But the tissue immediately below the integument is in an attenuated con- 
dition, and the body of the nymph is becoming vaguely outlined at a lower level, 
although it is not yet delimited by a definite cell wall. The cuticula intima of 
the rectum of the larva has been expelled, but remains attached externally to the 
body in the position of tlif larval anus; and the exuviae of the terminal portions 



BY VERA IRWIN"-SMITH. 61 

of all the larval tracheae which open on the surface by spiracles are found simi- 
larly attached externally to the pronymph in the position of the larval spiracles 
(Text-fig. 23). The expulsion of these exuviae may be coincident with the se- 
paration of the larval ectoderm to form the puparium; but it is possible that 
they represent a later eedysis. 

The exact time occupied by the different stages of pupal development in the 
Stratiomyiidae is exceedingly difficult to determine, since the onset of pupation 
is not marked by any change in outward form of the larva, and larvae of the 
type of Metoponia ruhrieeps are normally so sluggish that immobility does not 
attract anj^ special attention. 

The larva found in the pronymph stage was last observed in movement on 
October 26th, and its immobility was first noticed on November 13th, when it was 
dissected, and, as already stated, found to be dead. 

In the case of the blowfly, Lowne, while explaining that the rate of develop- 
ment depends largely on temperature, estimates that the formation of the pro- 
nymph from the larva occupies approximately about three days, that the head is 
evaginated about the fourth day, and that the nymph is completely formed by the 
end of the fifth day. The stage of development reached by the pronymph of 
Metoponia ruhrieeps described here Vv^ould correspond, according to this calcula- 
tion, with that of a blowfly pupa three days old. 

The Tracheal System of the Pronymph. 

Unlike the Cyclorrhapha, the tracheal system of the adult Metoponia ruhri- 
eeps is not formed independently, but in direct relation with that of the la,rva. 
The larval vessels do not disintegrate, but remain functional at the metamorphosis, 
and persist to form the foundation of the pupal vessels. Although modified by 
the histolysis of some portions and the change in size and relative importance of 
others, and by the great increase in number and extent of smaller branches, the 
general plan of the larval system is still clearty distingaiishable, and can be traced 
through all subsequent stages to the adult fiy. 

In the pronymph the two great longitudinal trunks are functional only to 
the fourth abdominal segment (Text-fig. 49). Behind this they still retain the 
form of open vessels, extending down to the position of the old larval posterior 
spiracles; but they are much degenerated and their cuticular lining has been 
detached and expelled from the body in the region of the spiracles. Towards 
their posterior end they are already quite vestigial, and their outlines and the 
outlines of the posterior spiracles are only faintly discernible. 

On the outer and lower side of each primary (i.e., larval) trunk, a secondary 
main longitudinal trunk has been formed by the straightening and enlargement 
of the vessels which formed the segmental lateral loops in the larva,, and the 
-ends of the loops, where they connected with the primary trunk, have now be- 
come stout commissures, from which a cluster of fine traeheoles arises in each 
segment. Posterior to the fourth abdominal -segment, where the primary trunks 
a.re degenerating, the ends of the commissures attached to them are also beginning 
to degenerate, and show a distinct line of demarcation between persistent and vesti- 
gial portions. On these commissures, of the fifth, sixth, and seventh abdominal seg- 
ments, the clusters of traeheoles are more voluminous than they are further for- 
ward, and ramify closely through the body tissues in the region between the two 
trunks. In the eighth segment, only the anterior portion of the segmental lateral 



62 



STUDIES IK LIFE-HISTORIES OF AUSTRALIAN DIPTERA BRACHYCERA, 



loop persists; the remaining portion, which functioned with the primary trunk 
of the larva near the opening of the posterior spiracle, has degenerated com- 
pletely, and only the faintest trace of it can be seen. The persistent remnant 
ends blindly in a thickened knob-like end, from which fine tracheoles radiate in 
all directions (Text-fig. 23). 

It is in this posterior region that the most marked difference is found be- 
tween the tracheal systems of larva and pupa. The tracheation has become en- 
tirely recentralised. The larval system of vessels, sjDreading upwards and out- 
wards from the region of the air sac and posterior spiracles, has completely dis- 
appeared, and a new system of fine tracheoles has arisen, based on the stump-like 
ends of the secondary longitudinal trunks in the region of the small lateral 
spiracles of the seventh segment. 

There is an increase of the number of delicate side branches supplying the 
body tissues in each segment, but, as in the ease of the larva, these are formed 
entirely in connection with the secondary longitudinal trunks. The primary 
trunks are still almost devoid of branches. The dorsal transverse commissures, 




Text-figures 23-26. Pupal respiratory system. 

23. Ventral view of posterior end of pronymph, showing termination of 
tracheal system, (x 421. (On the left side of the figure only the main frame- 
work of the system is shown), c.i., expelled cuticula intima of primary longi- 
tudinal trunk; p.t., vestigial primary trunk; r., expelled cuticula intima of larval 
rectum; s.t., segmental loop, forming portion of secondary longitudinal trunk; 
S6, sixth abdominal spiracle; 'S7, seventh abdominal spiracle. 24. Anterior 
spiracle of pronymph. (x 100). 2.5. Anterior spiracle of nymph, (x 100). 26. 
Anterior spiracle of proimago, showing as a vestigial structure (v) in the pupa 
sheath (p.s.), above the fully formed imaginal spiracle (i) . (x 100). 



BY VERA IRWIN-SMITH. 



63 



%hich in the larva, connected the two trunks in each segment, are still found 
in the thoracic and first to fourth abdominal segments, but behind this they have 
disappeared, evidently in relation with the disintegration of the trunks them- 
selves in this region (Text-fig. 49). A pair of delicate branches is given off 
anteriorly from the mid-region of each transverse commissure, and at least one 
pair anteriorly and one pair posteriorly towards the sides of the commissure. 
These branches do not show up clearly, and cannot be followed to their termina- 
tions. 




Text-figure 27. Respiratory system of pronymph. 



Ventral view of anterior end, showing the principal tracheal vessels, (x 57). 
CI, C2, C3, dorsal transverse ccmmisgures of the three thoracic segments; C4, dor- 
sal commissure of the first abdominal segment; l^_ l^. l^, vessels to the legs of 
corresponding thoracic segments; m.c, metathoracic lateral commissure; m.v., 
metathoracic ventral vessel; ph., exuviated pharynx of lar\'a; p.t., primary longi- 
tudinal trunk; S2, S3, lateral segmental loops of meso- and metathorax; VI, V2, 
first and second ventral transverse commissures; w., vessel to wing. 



64 STUDIES IN LIFE-HISTORIES OF AUSTRALIAN DIPTER^i BRACHYCERA^ 

The very fine tube which, in the larva, extends out from the fork of the 
lateral segmental loop in each abdominal segment to the corresponding lateral 
spiracle, has given place to a stout branch in the same position; and the secondary 
lateral spiracles are already well formed and functional. The development and 
structure of these spiracles are of sufficient importance to require special con- 
sideration, and they will be described later in a separate section. In the seventh 
segment the spiracles are simpler in form, but they too are connected with the 
secondary longitudinal trunks by a stout branch on each side (Text-fig. 23, S7). 

The tracheal system of the thoracic region is much complicated and confused 
by the supply of tracheae to the invaginated head, but, with a little care, the 
system which supplies the thorax and its appendages can be distinguished and 
followed out. 

In the larva, the course of the lateral segmental loops of this region could 
not be traced. In the pronymph, closed loops are formed in the second and third 
segments, but in each ease the loops bend abruptly inwards, across the course of 
the primary tracheal trunks (Text-fig. 27, S2, 3). From the inward bend in 
the second segment the secondary trunk runs forward below and parallel with 
the primary trunk, and junctions with it just in front of the transverse com- 
missure of the mesothorax. A branch is given off from the inward apex of the 
loop in both meso-, and meta-thorax to supply the corresponding leg (12, 3). 
The branches for the two prothoracie legs are given off anteriorly from a trans- 
verse commissure which connects the two primary trunks ventrally behind the 
prothoracie spiracles (11) . The main branch for the wing on each side arises 
in the region of the junction of primary and secondary trunks (w). Th« short 
branch in which it originates forks into two, one passing directly outwards to the 
wing, the other directly inwards to supply with its branches the dorsal surface 
of the mesothorax. 

Behind this, just in front of the fork junction of raeso- and meta-thoracic 
lateral loops, a dorsal branch is given off from the outer side of the mesothoraeie 
loop, which also supplies the dorsal area of the mesothorax, and gives off two 
slender branches which extend outward towards the wing. From the fork of the 
metathoracic lateral commissure (m.c.) an extremely fine tracheole extends out- 
ward, apparently to the metathoracic spiracle; but this, if present, is too small 
and inconspicuous to be detected. Jtist behind tlie metathoracic commissure a 
long branch arises from the metathoracic lateral loop and extends inward, ven- 
trally, to supply the mid-ventral region of the metathorax {m.v.). 

The dorsal transverse commissures connecting the primary trunks in the 
meso- and metathorax are quite similar to those of the abdominal segments, and, 
like them, are arched forward (C 2, 3). But the transverse commissure of the 
prothorax is bent backwards, behind and below th". mesothoraeie commissure, 
being drawn inward, evidently', with the invaginated head (C 1). It gives off a 
pair of short stout branches which extend posteriorly apparently to the region 
of the optic discs; but it is not possible to determine exactly what internal organs 
are supplied by the various tracheae, since their stnicture is not at all clear in 
this specimen. A much finer pair of traeheoles arises close together from the 
same commissure in the midline, between the stout pair, and extends anteriorly 
a short distance. 

In addition to the three dorsal transverse commissures, two ventral trans- 
verse commissures are found in the thorax. The first is that which gives off the 



BY VERA IRWIN-SJdlTII. 65 

pair of tracheae supplying the prothoraeic legs. It junctions with the primary 
trunks at the point of origin of the first dorsal commissure, and appears to be 
identical with the ventral commissure v/hich was observed in the larva (Text-fig. 
1). The cluster of radiating branches arising at its root, in the larva, is repre- 
sented, in the pronymph, only by some short degenerating stumps attached to the 
base of the dorsal commissure on each side, and by a long branch which extends 
up into the larval head-plate and pharynx. This, too, appears to be disinte- 
grating, and on one side it has already lost its connection with the primaiy 
trunk. 

Two long branches are given off posteriorly from the commissure, between 
the pair to the legs. They extend back into the mesothorax, and terminate in 
its mid-ventral region m a cluster of fine threads on a rounded structure just . 
behind the optic disc, apparently a nerve ganglion. 

The second ventral commissure (V 2) is formed in connection, not with the 
primary but with the secondary trunks. It is attached to eaeli mesothoracic 
lateral loop a short distance behind the junction of the latter with the primary 
trunk. It is bent sharply backward, and gives off posteriorly, from its middle, 
a pair of long branches which extend into the mid-ventral region of the meta- 
thorax, and end in fine threads attached to a large oval ganglion, situated im- 
mediately behind that supplied by the pair of branches from the first ventral 
commissure. 

The anterior spiracles are simple orifices v\rhich open directly into the tracheal 
trunks, without the intervention of vestibule or stigmatic chamber (Text-fig. 24j. 
The prothoraeic stigmatic plate and complicated stigmatic apparatus of the larva 
are shed ndth the larval skin Avhieh forms the pupa case, and the new aperture 
appears to be formed just below the position of the larval aperture, in relation 
with a new terminal portion of the tracheal trunk. A sear on the integument im- 
mediately in front of the present spiracle seems to represent the position of the 
larval opening, and there is some evidence of a faintly outlined vestigial vessel 
connecting this with the tracheal trunk; but none of these structures are well de- 
fined, this portion of the body wall being in poor condition. A mass of granular 
tissue surrounding the new trunk probably indicates the imaginal disc from which 
it has been formed, and corresponds with the disc described by Lowne as the 
dorsal prothoraeic imaginal disc. The shed lining of the trunk from a previous 
eedysis is still attached externally to the stigmatic opening. • 

Although the new opening is situated behind the j)Osition of the larval open- 
ing, it is certainly prothoraeic, and not intersegmental. It would appear that the 
terminal portion of the pronymphal tracheal trunk occupies the same position as 
the small vessel given ofL" from the base of the stigmatic chamber in the larva, and 
it may be formed in direct relation with this, as was pointed out in the dis- 
cussion of the larval prothoraeic spiracles. 

The whole structure seems to be very poorly developed, and of little im- 
portance for respiration at this stage. 

The Tracheal System of the Nymph. 

Two specimens of an early nymphal stage have been examined, the first 
with the head only half everted, the second (figured in part i. of this series, 
Irwin-Smith, 1920, Plate 27, figs. 3, 4) with head fully exposed, but with the 
optic stalks passing back through the neck to retinal discs which are still par^ 
tially enclosed witliin the thorax. 



66 STtfDIES IN LIFE-HISTORIES OF AUSTRALIAN DIPTERA BRACHYCERA^ 

The first is only slightly more advanced in development than the pronymph 
described ahove, and probably corresponds to the fourth day stage of Lowne's 
blowfly pupa. The shed lining, of the larval rectum is still attached externally 
to the. integument, but there is now no trace of the shed tracheal endings of the 
larva. The pupa sheath is not yet separated off, but the abdominal spiracles are 
now entirely on the surface and form prominent projections on the body wall. 
The tv/o prothoraeic spiracles also project slightly, the portion of the body 
waU surrounding them being raised into a rounded protuberance on each side 
of the neck (Test-fig. 25) but the protuberances are quite soft, without any 
strengthening of horny chitinous tissue. 

In the second specimen two distinct layers of integument are visible cover- 
ing the body, but the outer one is still closely applied to the inner and does 
not form a separated sheath. All trace of larval rectum and posterior larval 
spiracles has disappeared, and the external apertures of alimentary canal and 
tracheal system are now entirely pupal in character. Unfortunately the definition 
of the tracheal system in this specimen is not as clear as in the pronymph, and 
only some of the more prominent features are visible. These have just the same 
form as in the pronymph, but the main trunks and branches have become en- 
larged and thickened, and there rppears to be some increase in the number of 
finer branches. The protuberances on which the prothoraeic spiracles are 
situated have become more prominent; but they are still quite soft, and the 
spiracles are not provided with a horny cornu, or a digitate ornamentation, and 
appear to be only vestigial structures. Unlike the abdominal spiracles they are 
not, apparently, in communication with the external air through the pupa case, 
and it is uncertain whether they possess a lumen at this stage. Metathoracic 
spiracles, if present, are still quite unimportant, for neither their apertures, nor 
tracheal branches leading to them can be detected. On the other hand, the 
abdominal spiracles are highly developed, and of complicated character. Their 
development and structure can now be dealt with in detail. 

Development of the secondary abdominal spiracles of the pupa. 

As has been already stated, the development of the pupal spiracles begins in 
the larva some considerable time before pupation, and is marked in successive 
lai-val instars by changes in the form of the larval skin immediately above the 
developing spiracles. 

Satisfactory sections could not be obtained through this region owing to the 
coarse, tough, spiny nature of the cuticle, and the heavy deposition of chitin, 
so that it has not been possible to follow out, in histological detail, the earlier 
processes in their formation. But some of the structure underlying the cuticle 
can be made out in well cleared fragments dissected from the body of the larva. 

In the earlier stages, a mass of granular cells is visible in the vicinity of 
the primary larval spiracle, and at the fourth larval stage, represented by Text- 
fig. 19, this region has developed into the form of a cone having a blunt, inverted 
top, which projects up immediately below the spiny ridge on the cuticle, to one 
side of the primary spiracle (Text-fig. 28). It seems safe, then, to assume that 
the pupal spiracle is developed in the normal way, as an invagination in an 
imaginal disc, which adjoins or surrounds the termination of the primary larval 
trachea. Its homy tip must be everted when the larval skin is shed to form 
the pupa case, for it is found penetrating the case at the onset of pupation, and 
is then already open and functional. 



BY VERA IKWIN-SMITH. 



67 



in tbij prouymph (in which the head is still invaginated), the whole lateral 
stigir.atic apparatus of each abdonii^-al segment from th.e first to the '^.isth, with 




Text-figures 38-36. Development of abdominal spiracles of the pupa. 

28. Optical section through region of spiracle and fourth stage papilla of 
larva, (x 290). 29. Spiracle of pronymph. (x 140). 1., shed lining of larval 
spiracle; s., secondary spiracle developing in encapsuled disc. 30. Spiracle of 
early nymph stage, projecting above surface of body; p., integument of pronymph; 
n., integument of nymph. (x 125). 31. Fully developed spiracle, in pupa sheath 
of proimago. (x 140). f., chitinous funnel; 1., larval stigma. 32. The same, 
viewed from the inner side, (x 140). 33. Knobby processes on skin over funnel. 
(x 475). 34a, 34b, Two views of spiracular orifice. (x 290). 35. Portion of 
pupa sheath removed from puparium, with spiracles still attached, (x 48). sp. 6, 
sp. 7, spiracles of sixth and seventh abdominal segments; pr., processes on pupa 
sheath. 36. The same processes. (x 475). 



its surrounding disc, is found enclosed in a definite membranous capsule, deeply 
embedded in the somewhat attenuated tissue at the side of the body, and covered 



68 



STUDIES IN LIFE-HISTOBIES OF AUSTRALIAN DIPTERA BRACHYCERA^ 



by the integument, through which the horny tip of the pupal spiracle projects 
(Test-fig. 37). The portion of the trachea which enters the capsule is differen- 
tiated from the estra capsular portion by a slight constriction, and an apparent 
difference in composition. Within the capsule the stigmatic chamber of the 
pupal spiracle is seen in process of formation in the middle of the disc, and 
appears clearly as a secondary formation growing up round the terminal portion 
of the primary larval trachea (Text-fig. 29). It is still quite small, its maximum 
diameter no greater than that of the trachea where it enters the capsule. The 
terminal portion of the larval trachea persists as a delicate, thin-walled tube 
given off from its base, communicating vvith the larval spiracle on the integument, 
to the external surface of which its shed lining is still attached. It is apparently 
reformed after the eedysis, and persists, and, as far as can be ascertained, re- 
mains open throughout the whole pupal period. 

In the early nymphal period, when the head is just evaginated, the body is 
delimited by a new wall at a lower level than the old integument, and in the 
same plane as the base of the stigmatic capsules. The pronymphal integument 
is at first drawn back to this wall and is closely adherent to it. At this stage 
(Text-fig. 30) the stigmatic capsules appear as prominent projections on the 
surface of the body, but they are still covered by the old integument (p), which 
is strengthened by a deposition of chitin in the region surrounding the elongated 
neck of the spiracle, and forms a collar for it. The stigmatic chamber has now 
widened out to its characteristic pear shape, and the whole capsule has become 
much larger. 

Later the old integument is shed to form a distinct pupa sheath separated 
from the body by a considerable space. In this space the lateral spiracles come 
to lie, as already, explained (Irwin-Smith, 1920), entirely outside the body, each 
connected with it only by the tracheal vessel which supplies it, and attached to 
the pupa sheath and pupa ease (old larval skin) by the neck, which penetrates 
through them to the external surface (Text-figs. 31 and 38). 

As the nymph develops into the proimago, a septum is formed in the wall 
of the connecting trachea, below the surface of the body. This is strengthened 
])y a chitinous rim, and eventually forms the vestibule of the imaginal spiracle. 
When the. adult fly emerges, the trachea which connects the pupal spiracle with 
the body is severed at this point, and the whole spiraeular apparatus is left 
behind in the pupa sheath, enclosed in the pupa ease. If the sheath is pulled 
away from the puparium the spiracles are Avithdrawn from the perforations in 
the walls of the latter, and remain in position in the sheath, with the neck of 
each protruding through its protecting chitinous collar (Text-fig. 35), It is in 
this position, in the thin transparent ' sheath, removed from the puparium, that 
tlie structure of the fully developed spiracle can be studied most easily. 

Secondary spiracles of the pupa. When completely developed, the pupal 
spiracles have the form shown in Text-figs. 31, 32. 

Each is a large flask-shaped body, with an elongated slender neck, through 
which it opens on the outer surface of the puparium. The neck is stiff and 
horny and dark yellowish brown in colour, being highly chitinised. The stig- 
matic chamber into which it leads, opens at its base into the trachea connecting 
it with the body of the pupa. From the base a narrow delicate tube extends 
out to an aperture in the pupa sheath corresponding in position with the primary 
lateral spiracle of the larva. As already stated, this is the vestigial larval 
trachea, round which the pupal stigmatic chamber has developed. Evidence of 



BY VERA IRWIX-SMITH. 69 

the manner of its growth is seen in the shape of the chamber. Its walls form 
two enveloping- flaps, partially sheathing the primary trachea throughout tlie 
greater part of its length. The deep groove or crevice in it, which is formed 
in this way, extends right up the chamber and along the neck to the orifice, so 
that it is open on one side, the side nearest to the larval spiracle. 

The structure of the aperture is shown in Text-figs. 34a, Sib. Some half 
dozen wedge-shaped ribs of the ehitinous wall project into the opening, between 
oval perforations. The floor of the neck, from the orifice to the stiginatic 
chamber, bears a row of fine, stiff hairs, which slant uniformly upwards and out- 
wards towards the opening. 

The stigmatic chamber was described erroneously in my fii-st paper (1920) 
as having muscular walls. More careful examination shows a structure very 
similar in appearance to the prothoracic stigmatic chamber of the larva, in which 
the interior of the chamber is filled with a mass of chitin rods. I have not 
been able to section it, but from surface view under high magnification the rods 
are plainly Adsibie, embedded in a kind of matrix composed of what appears to 
be a fine reticulum. 

Measurements of the pupal stigmata are: — Total length from base of cham- 
ber to tip of neck .240 to .250 mm., length of neek .080 to .092 mm., diameter 
across neek . 025 mm., diameter across base of chamber . 062 to . 077 mm., length 
of primary larval trachea from junction with stigmatic chamber .129 to .148 
mm. 

The ehitinous collar for the neck spreads out widely below to form the pro- 
tecting "funnel," .085 mm. in depth and .148 mm. in maximum diameter, which 
covers the upper part of the stigmatic chamber (Text-fig. 32, /). Above the 
funnel, the surface layer of the thin pupa sheath is studded with numerous small 
"knobby processes, T\'hich seem designed to afford the chamber still further pro- 
tection (Text-fig. 33). 

The Tracheal System of the Pro-imago. 

At the latest nymphal stage^ the pupa sheath exists as a separate envelope 
only over the head and thorax and their appendages. A reddish-brown pigm.ent 
is appearing in the eyes, the legs have unfolded and extend down over the abdo- 
men, and the separate segments of legs and antennae and the articulations be- 
tween them are well defined, but the segments are all quite colourless; integu- 
mental hairs are faintly visible, and the pads and claws of the tarsi are forming. 
The abdomen is much less advanced in development, and does not show any trace 
of integumental hairs or setae. 

The tracheal system at this stage is very difiicult to trace, but appears stiU 
to retain the early nymphal form. Metathoracic spiracles are not visible, but 
the prothoracic spiracular humps are still protuberant, and the lateral abdominal 
spiracles are as described for the njrmph stage; 

The development of the proimago must proceed rapidly after the entire 
separation of the pupa sheath, for all the pupae examined at a stage subsequent 
to this are well advanced, and differ but little in external form from the imago. 
A pupa at this stage is figured in Part i. of this series (Irwin-Smith, 1920," PI- 
27, fig. 6). ^ 

The ehitinous exoskeleton of the imago is well developed, and deeply pig- 
mented, the body segments have the form found in the adult fly, and the com- 



70 STUDIES IN LIFE-HISTORIES OF AUSTRALIAN DIPTERA BRACHYCERA, 

plete ornamentation of integumental spines and hairs is present. An ornamen- 
tation of minute spiny processes is found also on the east pupa sheath, in the 
form of a rectangular patch on the lateral margins of each segment a little 
below the spiracles. The processes are arranged in groups of twos or threes in 
fairly even rows (Text-figs. 35, 36), and are similar to the processes found in a 
corresponding position on the body of the imago; but no other imaginal hairs 
or setae are repeated on the pupa sheath.* 

The tracheal system of the proimago is quite imaginal in character. It is 
already strongly developed, has become very voluminous, and is quite con- 
spicuous in cleared specimens, except where the more deeply pigmented portions 
of the exoskeleton of head and thorax obscure its finer ramifications. 

The most notable difference from the system in the nymph is found in the 
formation of large posterior thoracic spiracles and the development of an exten- 
sive system of large tracheal trunks in connection with them, so that this region 
becomes the principal source of air supply to the imago. The anterior thoracic 
spiracles of the imago are also fully developed, and the tracheal system with 
which they communicate has become greatly enlarged. 

The anterior spiracles are situated in the antero-lateral region of the thorax, 
on a level with the surface of the body. The prominent spiracular humps of the 
nymph have been separated from the body, and show only as vestigial struc- 
tures in the pupa sheath, with a faint trace of a shed tracheal tube attached to 
the lower border of each, immediately above the imaginal spiracle (Text-fig. 26). 
There seems to be no doubt that the thoracic spiracles of the imago are formed 
in direct relation with those of the larva, and that they are, therefore, definitely 
prothoracic and metathoracic in origin. The large tracheal trunk, into which 
each posterior spiracle opens, connects it with the primary and secondary longi- 
tudinal trunks in the exact position of the fine branch which, in the larva, and 
pronymph, communicates with the metathoracic spiracle. 

The pupal abdominal spiracles of the proimago have been described already. 
The corresponding imaginal spiracles are now well developed, and the tracheae 
which connect them with the pupal spiracles are very lightly attached, so that 
they are easily broken away when the pupa is being handled (Text-fig. 38). 

The tracheal system still shows the same general gToundplan as in the pro- 
nymph, and in the abdomen it is very little altered. The tw^) main longitudinal 
trunks, primary and secondary, are still present on each side. The secondary 
trunks have increased in importance and give off numerous side branches all 
along their course, but especially in the region of the commissures which connect 
them with the spiracles and the primary trunks. These commissures are now 
wide vessels. The primaiy trunks remain simple and practically unbranehed. 
Transverse commissures connecting the two primary trunks are present only in 
the second, third, and fourth abdominal segments. The commissure of the first 
abdominal segment, and the metathoracic commissure seem to have disappeared. 
I could find no trace of either in the proimagines and imagines examined. The 
primary trunks narrow abruptly just behind the fourth spiracles, and in the 
segments behind these are represented only by a delicate loop tube on each side, 

*This ornamentation of the sheath recalls that found on the first moult skin 
shed by the larva at hatching, which was described and figured in Part ii. of this 
series (Irwin-Smith, 1921, fig. 2) . The processes on the pupa sheath are, naturally, 
larger, but are of much the same form. 



BY VERA IRWIN-SMITH. 



71 



connecting the commissures from the secondary trunks (Text-fig. 50). The 
secondary trunks terminate abruptly in the seventh segment in a tuft of fine 
branches extending down into the tail region. 

in the thorax, the original plan is greatly obscured by the extensive develop- 
ment of secondarv tracheae; but the large simple primary trunks are still present, 
communicating with the anterior thoracic spiracles. They are evidently the 
"paragastric trunks" described by Lowne. In place of the secondary trunks 
there is a thick bundle of fine tracheoles on each side. But in this and all other 
details the thoracic system is quite similar to that of the imago, and will be 
described at greater lengih in the description of the latter. 




Text-figures 37-39. Lateral abdominal region, showing development of imaginal 
tracheal system from pupal system. 

37. Pronymph, first two abdominal segments. (x 48). 38. Proimago, first 
three segments. (x 43). 39. Imago, the same segments. (x 43). d.c, dorsal 
transverse commissure; p.s., pupa sheath; p.t., primary longitudinal trunk; s.t., 
secondary longitudinal trunk; sp. 1, first abdominal spiracle. (In the proimago, 
the tracheal connection between the pupal and imaginal spiracles has bsen broken) ; 
w., body wall of imago. 



The Tracheal System of the Imago. 

The anterior thoracic spiracles are large oval openings, each bordered by a 
chitinous rim, which is fringed by long, close-set, dark hairs (Text-fig. 43). The 
vestibule into which it leads is a rounded chamber having on its inner surface 
numerous short setae arranged in comb-like groups with thickened, chitinous 
bases. It communicates with the trachea proper through a narrow neck, which 
apparently acts as a valve. No other valvular apparatus is discernible. 

The posterior thoracic spiracles, which are situated slightly below and in 



72 



STUDIES IN LIFE-HISTORIES OF AUSTRALIAN DIPTERA BRACHYCERA, 



front of the bases of the halteres, are also large openings, very similar in struc- 
ture to the anterior spiracles, but the comb-like structures in the vestibule are not 
so well developed, and the fringe of hairs round the opening is uneven, being 
longest and thickest on the postero-ventral border. The ventral rim bears, in 
addition, a single large seta (Text-fig. 47). 




Text-figures 40-44. Imaginal respiratory system. 

40. Tracheal system in thorax. Dorsal view; semi-diagrammatic, (x 36). 41. 
Ventral view of anterior portion of the same, (x 36). 42. Tracheae in. region of 
anterior spiracle. Dorsal view, with primary trunk removed, (p.t.). (x 57). 3. 
Anterior spiracle, (x 133). 44. Abdominal spiracle viewed through the side of 
body, (x 133). 44a. Surface view of same, (x 133). 

a.s., anterior thoracic spiracle; /I, tracheae to prothoracic leg; m,c., meta- 
thoracic lateral commissure; n., tracheae to neck and head; p.s., posterior thoracic 
spiracle; p.t., primary trunk; si, spiracle of first abdominal segment; s.t., second- 
ary trunk; x., mid-ventral tracheae; y., tracheae to lateral region. 



The abdominal spiracles. — A pair of spiracles is situated dorso-laterally, close 
to the anterior margin of each of the first seven abdominal segments. They lie 
in deep depressions in that part of the lateral integument which bears an orna- 
mentation of short, dark hairs. The depression forms the atrium of the 



BY VERA IRWI>r-SMITH. 



73 



•spiracle, and the hairy covering is continued a short distance into its cavity. At 
its inner extremity there is a thick dark rim of chitin, which marks the point 
of attachment of the tracheal tube of the pupal spiracle, and which forms the 
valve of the imaginal spiracle. It has the shape shown in Text-fig. 44, being 
continued dorsally into a pointed lever-like prolongation. It leads into a short 
vestibule, with thin transparent walls, which are not supported by the spiral 
thickening present in the trachea and are almost invisible. No inner valve is 
discernible at the junction of vestibule and trachea. 

Tracheae. — A notable feature of the tracheal system of the adult Metoponia 
is the absence of the great air-sacs which constitute such an important part of the 
system in the Muscidae and other actively flying Diptera. The tracheae are 
greatly enlarged in certain areas, but appear to consist throughout of unmodified, 
cylindrical vessels, which exhibit the typical spiral arrangement in their walls. 
The primary trunks (p.t.) are simple vessels of large calibre, which bend sharply 
in towards the median line from the anterior spiracles, and, a little behind the 
level of the spiracles, are connected by a short transverse commissure, which 
seems to represent the mesothoracic commissure of the larva and pronymph ( Text- 
fig. 40). Behind this, they diverge slightly and occupy much the same position 




Text-figures 45-47. Imaginal respiratory system. 

45. Latero-ventral view of tracheal system in region of posterior thoracic 
spiracle, (x 57). 46. Dorsal view of the same, (x 57). 47. Posterior thoracic 
spiracle, (x 133) . 

h., haltere; 12, 13, tracheae to meso- and metathoracic legs; l.v., latero-ventral 
tracheae; p.d., postero-dorsal tracheae; p.s., posterior spiracle; p.t., primary trunk; 
si, first abdominal spiracle; s.t., secondary trunk; v., ventral tracheae. 



as in the earlier stages, extending down dorso-laterally through thorax and abdo- 
men to the fourth abdominal segment (Text-fig. 50). In the abdomen the 



74 



STUDIES IN" LIFE-HISTORIES OF AUSTRALIAN DIPTERA BRACHYCERA^ 



tracheal system is exactly as described for the proimago. In the thorax the post- 
nymphal development of tracheae is characterised by a multiplication of fine 
tracheoles in faggot-like bundles, which, instead of single large vessels, take the 
place of the original single tracheae. As in the abdomen, the original groiind 



¥ 



48 





Text-figures 48-50 showing development of imaginal tracheal system from the 

larval system, (x 23). 

48. Larva. 49. Pronymph. 50. Imago. (Only ths body of imago is 
shown, without hiread or thoracic appendages) . 



jjlan of the system is still discernible: its main features are shown in the accom- 
panying figures (Text-figs. 40, 46). A short and very wide trunk (m.c.) ex- 
tends in from the metathoracic spiracle to the primary longitudinal trunk. The 



BY VERA IRAVIN-SMITH. 75 

segmental loop, forming part of the secondary longitudinal trunk, which con- 
nects this with the first abdominal spiracle (SI), is much enlarged, and dips down 
sharply towards the ventral surface, where it gives off the bundle of traeheoles 
to the third leg (L3). Just in front of this a massive bundle takes its origin 
from a swollen portion of the loop (v) and, after a short course, the individual 
traeheoles from it spread out fan-wise to supply the mid-ventral region of the 
thorax. The anterior segmental loop, which forms the most anterior portion of 
the secondary trunk, extends along the sides of the thorax in a deeply ventral 
position (s.t.). It originates in a wide vessel given off antero-ventrally from the 
metathoracic spiracular commissure, which soon breaks up into three bundles of 
traeheoles, the foremost forming the segmental loop, the next the tracheal bundle 
to the mesothoraeie leg (12), and the third a radiating system of traeheoles supply- 
ing the latero-ventral region of the mesothorax (l.v.). Close to its junction with 
the primary trunk, just behind the anterior spiracle (cls.), two bundles are given 
off from the secondary trunk, the inner of which bends down towards the mid- 
ventral line, where its vessels come into contact, if they do not actually junction 
with those of its fellow from the opposite side (x.). The outer bundle consists 
of very numerous traeheoles Avhich spread out to supply the side walls of the 
thorax between the anterior spiracle and the wing (y-)- This secondary trunk 
system forms the first of the four large gToups of bundles given off from the 
primary trunk at the same point just behind the anterior spiracle. The second 
is the bundle which supplies the prothoracie leg (11) . On its inner side, just 
before it enters the leg, it gives off a small cluster of traeheoles which meet with 
those from the opposite side (Text-fig. 41), and apparently represents the an- 
terior ventral commissure which was noted in the pronymph. The third group 
consists of a thick bundle which passes inward to the neck («.), and, with its 
fellow from the opposite side, forms the tracheal supply to the head. Several 
traeheoles pass straight across between these bundles below the neck, and form an 
anterior commissure which seems to represent the prothoracie dorsal commissure 
of the pronymph. The course of the tracheae in the head could not be followed 
out. The antero-dorsal area of the thorax is supplied by dense clusters of 
radiating traeheoles which arise dorsally from the same spiracular region of the 
main trunk. These are shown only on Text-fig. 50. The postero-dorsal region, 
including the scutum and seutellum, and most of the dorsal muscles, is supplied 
by similar clusters of traeheoles arising from the dorsal side of the large trachea 
connecting the metathoracic spiracle with the primary trunk. The bundles which 
are given off anteriorly extend straight forward, separate traeheoles from them 
passing out at right angles and forming a parallel series to the muscles all the 
way up the thorax (Text-fig. 46, p.d.). 

Summary. 

The following are the most important of the features of the respiratory 
system described in this paper : — 

1, The larva is peripneustie, both functionally and structurally, having spiracles 
on prothorax, metathorax, and on the first seven abdominal segments, and 
posterior spiracles opening internally in a preapical air chamber. The an- 
terior spiracles are large, the other lateral spiracles very minute. An ad- 
ditional pair of spiracles appears to be present on the head. 

2. The spiracles of the newly-hatched larva are exactly similar in number and 



76 STUDIES IN LIFE-HISTORIES OF AUSTRALIAN DIPTERA BRACHYCERA^ 

in form to those of the fully mature larva, but in the older larvae a papilla- 
like fold is developed behind each abdominal spiracle. 

3. These folds mark the position of the pupal spiracles, which begin to develop, 
in discs surrounding the larval spiracles, a considerable time before pupation. 

4. At pupation the neck of each pupal spiracle, from the 1st to the 6th 
abdominal, penetrates to the exterior of the larval skin through the papilla. 
It communicates internally with the tra,eheal system of the pupa through a 
complicated stigmatic chamber and a tracheal vessel which lie in the space 
between the body of the pupa and the puparium, and is functional through- 
out the whole puj)al period. 

5. TEe pupal stigmatic apparatus is detached, when the imago escapes, by the 
severance of the connecting trachea below the surface of the body, the 
ruptured end of the trachea forming the vestibule of the imaginal spiracle. 

6. The imaginal spiracles occur on prothorax, metathorax, and on abdominal 
segments 1-7; the prothoraeie spiracles are developed in immediate relation 

• with those of the pupa and larva, but the metathoraeic spiracles are not 
distinguishable in the pupa until the last stage before the emergence of the 

fly- 

7. The tracheae of the larva do not disintegrate, but continue to function, dur- 
ing the pupal metamorphosis, the tracheal system of the imago being formed 
by direct development from the larval system : the general plan of this can 
be traced through all intermediate stages, and is still clearly recognisable 
in the adult fly. 

8. In the larva a series of closed segmental loops is formed along each side 
of the body by the longitudinal junction of external branches from the 
main tracheal trunks. These loops become strongly developed in the pupa, 
and eventually form a secondary longitudinal trunk on each side parallel 
with the primary trunk, and connected with it and the spiracles by trans- 
verse commissures. 

9. The greatly increased growth of tracheal branches which occurs in the later 
stages of pupation is connected almost entirely with these secondary trunks 
and their commissures, and the primary trunks do not persist behind the 
fourth abdominal segment. 

10. There are three distinct stages in pupal metamorphosis, viz. : — pronymph, 
nymph, and proimago, marked by at least two, possibly three, eedyses, dur- 
ing which jDortions of the tracheal endings are shed, and distinct develop- 
mental changes occur in the respiratory system. 

11. The tracheal system of the imago is already well developed in the proimago, 
and is very conspicuous for some time before the emergence of the adult 

12. The tracheae of the imago retain the form of cylindrical vessels with a 
spiral arrangement of the intima. Thin-walled membranous air-sacs are not 
discernible. In the thorax, the imaginal development of the tracheal system 
is characterised by the formation of thick faggot-like bundles of fine 
tracheoles. 

Note on the comparative morphology of the spiraeular apparatus. 

De Meijere made a special study of the stigmata of larva and pupa in al 
large number of families of Diptera, and published the results in two papers 
(1895, 1902) which form the most impoitant contribution to our knowledge of 
this subject. 



BY VERA IRWI2Nr-SMITH. 77 

As a result of the special investigation, in the present paper, of the Stra- 
tiomyid stigmata, we are now in a position to consider these structures in the 
light of de Meijere's general account and discuss the conclusions which he 
reached. He has shown that the. anterior and posterior stigmata in all the 
Diptera larvae examined by him conform to the same general plan, i.e., that 
they all possess a more or less well developed stigmatic chamber, called by him 
the "Filzkammer," which communicates with the surface by a variable number 
of oval or rounded spots (Tupfelstigma), borne generally on short branches of 
the "Filzkammer" with enlarged ends (KJnospen) ; and a remnant of an older 
tracheal ending, which he terms "Narbenstrang," attached to the integument and 
to the base of the "Filzkammer" by an outer and inner stigma sear (Stigmen- 
narbe). These features are all recognisable in both the anterior and posterior 
spiracles of the Stratiomyid larva and the abdominal spiracles of the pupa. 

De Meijere's investigations were concerned mainly with the Nematoeera and 
the Cyclorrhapha,, and the structure of the abdominal spiracles of the Stratiomyid 
pupa has evidently not been observed by him or any other worker. But he has 
shown that, in the peripneustic pupae of the ISTematocera, the abdominal and pro- 
thoracic spiracles are alike and resemble those of the prothorax of the larva, 
whereas, in the more advanced Cyclorrhapha pupae, the abdominal spiracles 
have ceased to function, and the prothora,cic spiracles have become modified at 
the tip into horny cornua (Athemhorner), which break through the wall of the 
puparium to function in breathing. It is now evident that, in their mode of 
formation and general structure, the abdominal spiracles of the Stratiomyiidae 
show a close relationship with those of the Nematocera. They have a broad 
general resemblance to the abdominal spiracles of the Mycetophilid pupa, Bolito- 
phila cinerea Meig., which de Meijere figures as typical of the Nematocera. But 
the long, narrow, horny neck which perforates the puparium wall is distinctive, 
and recalls the prothoraeic stigmata of the Cyclorrhapha,. Other peculiar features 
of the abdominal spiracular appara,tiis of Metoponia are the great distance at 
which it is situated outside the. body of the pupa, the remarkable length of the 
tracheal trunk attaching it to the body, and the bulbous shape, large size, and 
complex development of the stigmatic chamber (Filzkammer). 

De Meijere rightlj^ remarks that, in the Stratiomyiidae, the prothoraeic stig- 
mata of the pupa are not specially developed, being represented only by weak- 
skinned colourless projections. And he records the pupae as peripneustic, but 
considers that the lateral stigmata are not connected with the body of the pupa 
by functional tracheae. Commenting on Brauer's statement that in the Stratio- 
myiidae and Cyclorrhapha the puparium case remains in vital connection with 
the pupa, he says "Diese Durchtrittssteilen sind aber die Stigmennarben, welehe 
bald fast ganz verschlossen sind; aueh sind die alten Tracheen so bald zusam- 
mengeschrumpft, dass ihnen fiir die Athmung der Puppe wohl, wenigstens in den 
spatern Stadien, keine Bedeutung beigemessen werden kann." But close in- 
vestigation has now shown that these abdominal spiracles form the only functional 
t)reathing apparatus of the pupa, and that they function in the same way as the 
prothoraeic stigmatic cornua of the Cyclorrhapha. , 

With reference to the old tracheal ending round which the Filzkammer is 
formed, whether in the larva or pupa, de Meijere considers that it always falls 
together to form a solid cord, whence the name "Narbenstrang." But it is 



78 STUDIES IN LIFE-HISTOEIES OF AUSTRALIAN DIPTERA BRACHYCEKA^ 

doubtful if tliis is so in the Stratiomyiidae. It has been shown that the cuticula 
intima of these tubes is shed in an ecdysis which occurs after the stigniatie 
chamber is fully formed, so that they must possess a lumen up to this time. And 
even after it they have not the appearance of solid cords, nor do their ends 
appear to be overgrown to form closed stigma scars. 

Special interest attaches to the perforation of the wall of the puparium by 
the horny neck of the abdominal spiracles. 

De Meijere devotes much attention to this ''Durchbruch" in the case of the 
Cyclorrhapha, but makes the definite statement that it does not occur outside this 
group. "Es kommt dies nur bei Cyelorrhaphen vor, indem bei den Stratiomyiden 
(der einzigen Familie der Orthorrhaphen, wo es ebenfalls zur Bildung eines 
Pupariums kommt) die Prothorakalstigmen auf nur sehr wenig hervortretenden, 
eonisehen Warzen aufsitzen." His mistake is due to the fact that he had over- 
looked the abdominal spiracles of the Stratiomyiidae, and was considering cmly 
the feebly developed prominences which, in this family, represent the stigmatic 
horns of the prothorax. In the Cyclorrhapha the perforation of the puparium 
wall is shown by de Meijere to occur on different segments of the pupa in 
different genera, e.g., on the 1st abdominal segment in many Syrphidae, and on 
the 2nd abdominal segment in the Phoridae; but in all cases only a single pair 
of horns is developed, which belong to the prothoracie spiracles. 

It is evident from the observations of de Meijere himself, and other authors 
whom he quotes, that the spots at which the perforations will afterwards occur 
are already distinctly observable on the larval skin in some, if not all, Cyclor- 
rhapha, just as they are in Metoponia. As long ago as 1740 Reaumur had no- 
ticed these spots on Eristalis tenax L., and described them as light circles on a 
dark ground, closed only with a thin hyaline membrane. And he recognised and 
described the pupal horns in process of formation below them, and stated that 
the time which elapses, after the formation of the puparium, before the horns 
perforate the spots, varies from 24 hours to 3 or 4 days, according to the lime 
of the year. Perris (1870) witnessed, in Xylota pigra F., the actual process of 
perforation, and found that it only occupied a few seconds. 

This, then, is clearly analogous to what occurs on the abdominal segments 
of Metoponia, and the study of the latter has revealed, in a very striking way, 
the intermediate position which the Stratiomyiidae occupy between the Nema- 
toeera and the Cyclorrhapha. The complete development of compound lateral 
stigmata links them, on the one hand, wi1?li the most primitive of the Nematoeera, 
and the modification of these structures, to provide for an air supply to a- 
"mummified" pupa enclosed in a puparium, indicates the beginning of a process 
which is carried to its highest development in the Cyclorrhapha. 

De Meijere discusses the problem of how the special places on the larval 
skin could have arisen in the Cyclorrhapha, and become adapted to the part they 
play in the thrusting forth of the stigmatic horns. He offers the tentative sug- 
gestion that the horns first succeeded in breaking through in species which pos- 
sessed a relatively thin wall to the puparium, and that, as the wall gradually 
hardened, the development of the imago was successful only in those which still 
preserved thin spots. In the Stratiomyiidae we have found the family in which, 
apparently, the adaptation originated, and it is here that its development can be 
studied most profitably. 



BY VERA IRWIN-SMITH. 79 

Bibliography. ; 

Papers wliieli I have not had an opportunity of seeing are marked with an 

asterisk. 

Austen, E. E., 1899. — On the preliminary stages and mode of escape of the 
imago of the Dipterous genus Xylomyia Eond. {Subula Mg. et 
auct.) with special reference to Xylomyia maculata F., and on 
the systematic position of the genus. Ann. Mag. Nat. Hist., 
(vii.), iii., pp. 181-190. 

Brauer, F., 1883. — Die Zweiflugier des Kaiserlichen Museums zu Wien. III. 
Systematische Studien auf Grundlage der Dipterenlarven, etc. 
DenJc. k. Akad. Wiss. Wien. Math. Nat. CI., xlvii., pp. 1-100, 
Pis. i.-iv. ( Stratiomyiidae, pp. 22-23). 

BrowN; J. M., 1910.^ — Some points in the anatomy of the larva of Tipula maxima. 
A contribution to our knowledge of the respiration and circula- 
tion in insects. Trans. Linn. Soc. London, xi., pt. 7, 2nd ser., 
pp. 125-135, Pis. 24-27. 

Carpenter, G. H., and Pollard, F. J. S., 1918. — The presence of lateral spiracles 
in the larva of Hypoderma. Proc. Boy. Irish Acad., xxxiv., Sec. 
B, No. 4, pp. 73-84, Pis. viii.-xiii. 

* Dewitz, 1890. — Einige Betrachtungen betreifend das geschlossene Tracheen- 

system bei Insectenlarven. Zool. Anz., xiii., p. 500. 

Enderlein, 1899. — Die Kespirationsorgane der Gastriden. Sitz. k. Akad. Wiss. 
Wien., Math.-Nat. CL, 108, Abth. 1, pp. 235-303. 

* , 1901. — Ueber die Gattung Gyrostigma Brauer, und Gyro- 
stigma conjungens, n.sp., nebst Bermerkungen zur Physiologie. 
Arch. Naturg., Ixvii., Beiheft, pp. 23-40, PL 1. (Stigmata of 
Dipterous larvae, physiology of respiration). 

Froggatt, J. L., 1918. — A study of the external breathing apparatus of the larvae 
of some Muscoid flies. Proc. Linn. Soc. N.8.W., xliii., 3, pp. 
658-667, PI. Ixx. 

* Geestaecker, a., 1874. — Ucber das Vorkommen von Tracheenkiemen bei aus- 

gebildeten Insecten. Zeitsch. tviss. Zool., xxiv. 

Halidat, a. H., 1857. — On some remaining blanks in the natural history of the 
native Diptera (larvae), etc. Nat. His. Bev., iv., pp. 177-196. 
(Stratiomyiidae, pp. 192-194). 

Hart, C. A., 1895. — On the entomology of the Illinois River and adjacent waters. 
Bull. III. State Lab. Nat. Hist., iv., 6, pp. 149-273. (Stratio- 
myiidae, pp. 247-266). 

Hewitt, C. Gordon, 1914. — The house-fly, Musca domestica Linn. Its structure, 
habits, development, relation to disease and control. Cambridge 
Univ. Press. (Respiratory system, pp. 41-47, 140-143). 

Irwin-Smith, V., 1920. — Studies in life-histories of Australian Diptera Brachy- 
eera. Part 1. Stratiomyiidae. No. 1. Metoponia rubriceps 
Macquart. Proc. Linn. Soc. N.S.W., xlv., pp. 505-530, Pis. 
xxvii.-xxviii. 

, 1921. — Ditto, No. 2. Further experiments in the rearing of 

Metoponia rubriceps. Proc. Linn. Soc. N.S.W., xlvi., pp. 252- 
255. 



80 STUDIES IN LIFE-HISTORIES OF AUSTRALIAN DIPTERA BRACHYCEKA^ 

-, 1921.— Ditto, No. 3. On the structure of the mouth parts 



and pharynx of the larval Metoponia rubriceps. Proc. Linn. 
Soc. N.S.W., xlvi., pp. 425-432, PI. sxxiii. 

JUSBASCHJANZ, S., 1910. — Zur Kenntniss der naeh embryonalen Entwickelung der 
Stratiomyiden. Jenaische Zeitsch., xlvi., pp. 681-736, 3 Taf. 

Krancher, 0., 1881. — Der Bau der Stigmen bei den Insekten. Zeitsch. wiss. 
Zool., XXXV., pp. 505-571. (Diptera, pp. 525-542). 

LowNE, B., 1890-1895. — The anatomy, physiology, morphology, and development 
of the Blow-fly {CalUphora erythrocephala) . A study in the 
comparative anatomy and morphology of insects. 2 vols. 
London. (Respiratory system of larva, I., pp. 47-50, of pupa, 
I., pp. 306-307, 319, 338-340; of imago, I, pp. 180-181, 224-227, 
II., pp. 351-368; development of tracheal system, II., pp. 368- 
373; physiology of respiration, II., pp. 373-386). 

LuKDBECK^ Wm., 1907. — Diptera Danica, Pt. 1, Stratiomyiidae, etc., Copenhagen. 
London. (Respiratory system of larva, p. 15). 

MagGregor, M. E., 1914. — The posterior stigmata of dipterous larvae as a diag- 
nostic character: etc. Parasitol., vii., pp. 176-188, Pis. xv., xvi. 

Mallock, J. R., 1917.^ — A preliminary classification of Diptera, exclusive of 
Pupipara, based upon larval and pupal characters, with keys 
to imagines m certain families. Pt. 1. Bull. Illinois State Lab. 
Nat. Hist., xii., 3, pp. J 61-407, Pis. xxviii.-lvii. (Stratiomyiidae, 
pp. 315-346). 

Meijerb^ J. C. H. de, 1895. — Ueber zusammengesetzte Stigmen bei Dipteren- 
larven, nebst einem Beitrag zur Metamorphose von Hydromyza 
livens. Tijdschr. Entom., xxxviii., pp. 65-100, 33 text figs. 
(Orthorrhapha, pp. 85-90). 

, 1902. — Euber die Prothorakalstigmen der Dipteren-puppen. 

Zool. Jahrb. Anat., xv., pp. 623-693, Pis. xxxii.-xxxv. (Stratio- 
myiidae, pp. 639-640, 676, figs. 33-34). 

, 1916. — Beitrage zur Kenntniss der Dipteren-Larven und. Pup- 
pen. Zool. Jahrb., Jena. Abt. System. 40, 177-322, Pi. iv.-xiv. 
(Stigmata, pp. 278-282.) 

Metcalp, C. L., 1919. — A proposed nomenclature for the parts of the posterior 
respiratory apparatus of Dipterous larvae, and a micro-protrac- 
tor useful in their description. Psyche, xxvi., 53-58, PI. i. 

* Palmen, J. A., 1877. — Zur Morphologie des Tracheensystems. 

■* Pantel, J., 1901. — Sur quelques Details de I'Appareil respiratoire et de ses An- 
nexes dans les larves de Muscides. 1. Stigmates et Trachees 
rudimentaires. Bull. Soc. entom. France, pp. 57-8. 

, 1909. — Sur les organes rudimentaires des larves des Mus- 
cides. Comp. Bend. Acad. Sci. Paris, 148, pp. 107-110. (Ap- 
Y^arcil respiratoire, pp. 109-110). 

Patton and Craig, 1913. — A text book of Medical Entomology. Lond. and 
Madras. (Diptera, pp. 8-308: Anat. and physiol., pp. 8-150; 
respiratory system, 96-100, PI. xx.). 

PORTIER^ P., 1909. — Physiologie de I'appareil respiratoire des larves d'oestres. 
C.B. Soc. biol., Paris, Ixvii., pp. 568-571. 

Pkennakt^ a., 1899. — Terminaison intracellulaire et reellement cytoplasmique des 
trachees chez la larve d'oestrc du cheval. C.B. Soc. biol. Paris, 
pp. 507-510. 



BY VERA IRWIN-SMITH. 81 

* ^ 1900. — Notes eytologiques, cellules traeheoles des oestres. 

Areh. d'Anat. Micr., iii. 

* SOSNOWSKI^ J., 1902. — Contribution a I'etude de la pliysiologie du develo]:^pe- 

ment des mouehes. Bull. Ac. Cracovie, pp. 568-573. (Sum- 
mary, Journ. R. Micr. Soc, 1903, p. 172). 

Tragardh^ I., 1914. — Skogsentomologiska bidrag 1-5. Pachijgaster minutisaima 
Zett., en under bark levende Stratiomyid. Entom. Tidskrift, 
Uppsala, XXXV., pp. 192-19G, figs. 3-5. (English summary, p. 
208). 

Vaney, M., 1902. — Contributions a I'etude des larves et des metamorphoses des 
Dipteres. Theses I'Universite de Lyon. No. 30, {Ann. Univ. 
Lyon., N.S. 1, Fase. 9), 178 pp., 4 i:)lates. (Appareil respira- 
toire, pp. 122-140). 

ViALLANES^ M. H., 1882. — Recherehes sur I'Histologie des Inseetes, et sur les 
phenomenes histologiques qui accompagnent le developpement 
post-embryonnaire de ces animaux. Ami. Sci. Nat. (Zool.), 
xiv., 1, pp. 1-348, PI. iv. (Histolyse des trachees, pp. 177-187; 
Histogenese des trachees, pp. 246-249). 

, 1882. — On the post-embryonic development of the Diptera. 

Ann. Nat. Hist. (5), ix., pp. 61-63. 

* VOGLER^ C. H., 1887. — Die Traeheenkiemen der Simulienpuppen. Mitth. 

Sclnoeiz. entomol. Ges., vii. 

* , 1900. — Beitrage zur Metamorphose der Teicliomyza fiisca. 

Illustr. Zeitschr. Entomol., v., pp. 1-4, 17-20, 33-36. (Respira- 
tory and oral structures of early stages of Teichomyza). 

' , 1900. — Weitere Beitrage zur Kenntnis von Dipteren-Larven. 

Ibid., pp. 273-276, 289-291. (Respiratory structures of Dip- 
terous larvae). 

* Wahl, Bruno, 1899.- — Ueber das Traeheensystem und die Imaginalscheiben der 

Larve von Eristalis tenax L. Arh. Zool. Inst. Wien, xii. 

* Weismann, a., 1864. — Die Entwieklung der Dipteren. Ein Beitrag zur Ent> 

wicklungsgesehiehte der Insecten, mit 14 Kupfertafeln. i. Die 
I Entwieklung der Diptera im Ei. ii. Die Nachembryonale 
Entwieklung der Musciden. (Zeitschr. f. wiss. Zool., xiv., 
1864). 

* WiSTiNGHAUSEN^ 1890. — Ueber Tracheenendigningen in den Sericterien der Rau- 

pen. Zeitschr. f. iviss. Zool., xlix., pp. 565-582, taf. xxvii. 

Postscript, added April 18th, 1923. — From larvae collected on November 4th, 
1922, showing fourth stage papillae, a male Metoponia ruhriceps emerged on 
April 4th, and a female on April 9th, 1923, and the three remaining larvae have 
now pupated. The male lived for eight days and the female for five. 



82 



STUDIES IN SYMBIOSIS. 

III. CONTRIBUTIOOSr TO THE MORPHOLOGY AND PHYSIOLOGY OF THE 
ROOT-NODULES OF PODOCARPUS SPINULOSA AND P. ELATA. 

By John McLuekie, M.A., D.Sc, Lecturer in Plant Physiology, University of 

Sydney. 

(With twenty-one Text figures.) 

[Read 28th March, 1923.] 

Introduction. 

Recent research has extended our knowledge of organisms capable of utilising 
the free atmospheric nitrogen. It is now known that plants having affinities with 
groups other than the Leguminosae have this faculty. Fixation of free nitrogen 
has been definitely proved to occur in the Cycadaceae, in the Podocarpineae, in 
Alnus, Eleagnus, and Myrica gale. All these plants possess some form of root-, 
nodules or tubercles which are inhabited by an organism capable of nitrogen fixa- 
tion. Bottomley (1907) isolated from the nodules of Alnws and Eleagnus a 
baciUus which, in pure culture, he employed to inoculate the roots of Vicia sativa 
seedlings. The result of these cross-inoculation experiments was the formation of 
root-nodules characteristic of those formed in nature by thp organism Pseudo- 
monas radicicola upon the roots of Leguminous plants. 

In 1899, Nobbe and Hiltner proved that the nodules which occur upon 
Podocarpus roots have the power of nitrogen fixation. They demonstrated, by 
their experiments, (1) that it is impossible to cultivate Podocarpus in the absence 
of the organism which causes nodule formation; and (2) that Podocarpus may 
be cultivated in pure quartz sand, containing no combined nitrogen, to which only 
the atmospheric nitrogen had access. 

In the earlier investigations on Alnus, Eleagnus, and Podocarpus, hyphae- 
like structures were observed in the cells of the nodules, and it was claimed that 
tliese represent a myeorhizic fungus to which the power of nitrogen fixation could 
be attributed. The hyphae-like structures were frequently disorganised and 
digested by the root cells (Shibata, 1902). These conceptions, however, have not 
stood the test of the most recent research, and the investigations of Bottomley 
and Spratt Ipave no doubt about the bacterial origin of the root-nodules in these 
organisms. Spratt (1912) has demonstrated that Pseudomonas radicicola stimu- 
lates the nodule formation in Alnus and Eleagnus, and that bacterial infection is 
responsible for the development of the nodules in Podocarpus totara, P. cMlina, 
P. alpina, Microcachrys tetragona, Dacrydium Franhlini, Saxegothaea conspicua 
and Phyllocladus trichomanoides. 



BY J. MCLUCKIE. 



83 



In the present note the author records results of a detailed study of the 
nodules of two other species of Podocarpus, namely, P. spinulosa and P. elata, 
which occur in the New South Wales Flora. 

The Nodules of Podocarpus spinulosa. 
In Podocarpus spinulosa the nodules are small, practically spherical, numerous 
and arranged in two close rows along the sides of the root. In P. elata, the 
nodules are also spherical, but less numerous than in the former species (Text- 
figs, la, Ifo). Branching of the root-system is relatively infrequent. The nodules 
arise mainly upon the root-hair zone of the mother root, and each nodule jjossesses 
numerous, normal-looking root-hairs. In the young nodules these hairs are ap- 
parentl}^ functional as they contain protoplasm and a nucleus. On the older 
nodules the hairs have collapsed and become partly torn away, but their vestiges 





Text-figs, la, lb. — Portion of roots of Podocarpus spinulosa and P. elata showing 

the root-nodules, (x 2/3.) 
Text-fig. 2. — ^Surface view of a root-nodule of P- spinulosa showing the root-hairs 

and the hyphae of a fungus (f). (x 33.) 

are apparent. In the species I have examined, the hairs upon the nodules are 
not so numerous as described in other species. In a nodule which has just reached 
maturity the hairs do not develop simultaneously, and hairs of various stages may 
generally be seen. They arise from ordinary epidermal cells of the nodule and 
are similar in all respects to normal root-hairs. Around each nodule and along 
practically the entire surface of the main root, there is a comparatively dense 
tangle of fine, fungal hyphae, some of which undoubtedly enter the root and, fre- 
quently, the nodule. 

Each nodule is endogenous in origin, and represents a modified lateral root, 
arising from the pericycle. A small vascular strand enters the nodule, and is con- 
nected with the vascular tissue of the main root, while the surrounding cortical 
cells of the nodule are enlarged and thin-walled. Many of these cortical cells are 
filled with wide hyphae-like threads traversing the cell cavities in different direc- 
tions. These have been interpreted by early observers as the hj^phae of a myeo- 
rhizic fungus. Sections stained in (a) Gentian violet, Lugol's iodine and safrauin, 
and in (&) Amyl gram stain (Kiskalt's) demonstrate that these threads are 



84 



STUDIES IN SYMBIOSIS^ 111., 



zoogloeal in nature and are occupied and formed by rod-shaped bacteria whieb 
occur in large numbers. The threads are of variable width, and the bacteria pass 
from cell to cell within them. The threads widen considerably before entering a 
new cell (Text-fig. 8a), much in the same way as Spratt has observed in other 
species of the Podocarpineae. In some cells, however, numerous small free 
colonies of bacteria occur outside the threads, and these divide independently of 
the organisms in the threads (Text-figs. 7, 8). 

In a few instances I have observed entire cells filled with these colonies of 
free individuals. 




Text-figs. 3, 4, 5.- 



-Longitudinal section of main root, through young root-organs. 
The nodules arise from the pericyle (p), and their early de- 
velopment is similar to that of normal lateral roots. In 
figures 4 and 5 the meristematic cells of the periblem are in- 
fe'ted by bateria (b) vhich enter from the cortex of the main 
root. (Figs. 3 and .5, x 30; Fig. 4 x 120.) 



BY J. MCLUCKIE. 



85 



The bacteria mass together round the nucleus when Iree, or the zoogloea 
threads coil round the nucleus; the latter again may undergo amitotic division in 
the presence of the bacteria. I have seen as many as four nuclei in an infected 
cell, but have not observed any stages in a normal mitosis. The shape of the 
nucleus also is very irregular as is shown in Text-fig. 21. In Text-fig. 21a, one 
of the nuclei is centrally constricted and apparently about to fragment. This 
irregularity in the form of the nucleus has been observed by Spratt (1912) in 
the nodules of Eleagnus and various members of the Podoearpineae, and by the 
author (1922) in the cells of Dipodium punctatum infected by an endophytic 
fungnis. The cytoplasm and nuclei of many of the nodule cells disintegrate and 
many of the bacteria leave the cells in which disintegration is taking place. These 
cells become strengthened by bars of thickening, which gives them a tracheidal- 
like appearance. Many of them are absolutely empty, but some contain the re- 
mains of zoogloea-threads, and numerous quiescent bacteria. The resting bacteria 
appear larger than the active individuals. 

Towards the end of the seasonal vegetative activity of the plant, the root- 
nodules become more or less completely transformed into masses of tracheidal- 
like cells, probably functioning later in water storage. These cells also appear in 
the cortex of the older portion of the main root. 




Text-fig. 6. — Longitudinal section of main root through mature nodule which is 
highly infected with the bacteria (b). Some of the cortical 
cells (c) have already been strengthened by the bars, while 
young root hairs are developing (h). (x 120.) 

In the spring new cortical tissue is formed within the old from the meriste- 
matic tissue within the endodermis. The outer, dead, cortical cells of the nodule 
consequently become crushed and disorganised. 

In Podocarpus spinidosa, the bacteria enter the root by means of the young 
root-hairs, and I have frequently seen young zoogloea-threads of the bacteria 
within the hairs passing inwards towards the cortex 91 the root. The bacteria 
multiply rapidly and numerous threads spread out in various directions into the 



86 



STUDIES IN SYMBIOSIS^ lU,, 



adjacent cells of the cortex. These zoogloea-threads in many cases look like 
fungal hyphae, but the bacteria can clearly be seen within them. In a very short 
time a considerable area of the cortex becomes infected. Threads of the bacteria 
spread into the young root origins and multiply within the meristematic cells. 
Text-figures 3, 4, 5, show the origin of the nodule from the pericycle (p), and 
the early period at which it may become infected by the bacteria (b). Infection 




Text-figs. 7, 8, 8a.— Cortical cells of the mature nodule showing zoogloea threads 
of the bacteria (th) more or less coiled in the vicinity of the 
nucleus or nuclei; (b) free bacteria; (n) nucleus undergoing 
amitotic division, (x 450.) 

apparently arrests meristematic activity, but stimulates the rapid enlargement of 
the cells. Infected cortical cells become larger than uninfected. The nodule 
gradually develops through the overlying cortical cells of the mother root, and 
appears as a more or less globular swelling on the surface. Text-figure 6 shows 
a mature nodule just after it has ruptured the overlying tissues of the main root 
and emerged from them. The bacteria enter embryonic cells of young root 
origins in Poclocurpus, and their presence leads to the cessation of cell division, 



BY J. MCLUCKIE. 



87 



but stimulates a rapiel enlargement of the young cells and, before the young root 
emerges from the cortex of the main root, many of its cortical cells are filled with 
bacteria which multiply rapidly, probably at the expense of the soluble carbohy- 
drates which are associated with evej-y active embryonic tissue. 

The active bacteria in some of the cells stain less deeply with gentian violet 
than the larger individuals in other cells which have lost their protoplasm and 
become transformed into reticulately thickened tracheidal-like cells. In fact, the 
two kinds of individuals, the active and inactive, are readily distinguished by 
means of this difference in reaction to stain. These larger organisms are pro- 
bably baeteroids. They are practically confined to the older cells of the nodules, 
in which disorganisation is taking place. The younger cortical cells of the nodule 
with dense granular cytoplasm contain the very characteristic zoogioea-threads 







Text-fig. 9. — Cortical cells from mature nodule with three nuclei (n) showing signs 

of disintegration, disorganising zoogloea threads (th) and 

resting bacteria (b) and the first thickening strands upon the 

cell wall (st). (x 720.) 
Text-fig. 10.— Small cortical cell containing the resting bacteria (bactsroids) and 

the disorganising threads and nucleus, (x 900.) 
Text-figs. 11, 12. — Cortical cells showing zoogloea threads of bacteria and two 

nuclei surrounded by dense granular cytoplasm, (x 600.) 
Text-figs. 13, 14.— Old cortical cells containing disorganising zoogloea threads and 

four nuclei from which the typical structure is disappearing. 

(x 600.) 

ot active bacteria, and there is no doubt that these active bacteria migTate from 
cell to cell in these threads, which very strongiy suffa'est the "infection threads" 
of the tubercles of the Legiiminosae (Text-figs. 7, 8). 



88 



STUDIES IN SYMBIOSIS^ 111., 



The threads almost invariably twine around the nucleus which, in the early 
infection stage, is in the reticulum condition. The cytoplasm of the cell is 
densely aggregated in the vicinity of the nucleus, which is the active chemical 
centre of the cell. I have not seen starch in these young infected cells, but since 
they occur in a zone just outside the endodermis of the nodule, and abundant 
starch is present in the cells of the main root at the base of the nodule, it is 
highly probable that the carbohydrate reaches the cells containing bacteria in the 
form of sugar. So long as the bacteria are active, i.e., during the vegetative 
activity of the host, they will consume carbohydrate which provides them with 
the necessary energy for nitrogen fixation, and there will be established the con- 
ditions required for a regular flow of carbohydrate from the tissues of the main 
root. 





Text-fig. 15. — T.S. of main root through a root-nodule showing the origin of the 
latter opposite a protoxylem group (p.x.) of main root, (x 
66.) 

Text-fig. 16.— T.S. nodule showing epidermis (ep), cortex (c), endodermis (end) 
and stele, (x 80.) 

Towards the end of the active vegetative period, the bacteria pass into a 
resting phase during which they appear larger, and stain more deeply (Text-figs. 
9, 10). The cortical cells of the nodules become transformed into water storage 
tissue by the disorganisation of the cytoplasm and nuclei and the deposition of 
thickening strands upon the wall. 

The single nucleus, in young infected cells, generally divides, and cells with 
2, 3, and 4 nuclei are common (Text-figs. 11 — 14). The chromatin contents be- 
come aggregated into small, irregular sized knots which rapidly disorganise. The 
zoogloea threads, at the same time, appear to disintegrate, and only small frag- 
ments of them may be evident. Many of these tracheidal-like cortical cells have 



BY J. MCLUCKIE. 89 

no resting bacteria in them, for the active bacteria, as a rule, migrate into the 
living epidermal and sub-epidermal cells of the nodule where they rest until the 
following season. 

Origin of the Root-Nodule. 

There is absolutely no doubt that the nodules are lateral roots, whose normal 
development has been arrested by bacterial infection. Text-figure 15 shows the 
origin of the nodule opposite one of the protoxylem groups of the diareh mother 
root. The connection of the central strand of traeheids of the nodule with the 
protoxylem group is clearly shown. Text-figures 3, 4, 5, show a longitudinal 
section of the main root through young root origins, and it is apparent that the 
pericycle cells become meristematic and divide to form a small swelling beneath 
the endodermis. At first the swelling is composed of a mass of small meriste- 
matic cells, with a large nucleus and granular cytoplasm, but, as the swelling 
develops, there are soon differentiated three distinct zones, namely, the plerome, 








17 18 

Text-fig. 17. — A group of the "coccus form" of the nodule bacteria showing thick 

wall and granular bodies, (x 2000.) 
Text-fig. 18. — A group of the "bacillus form" of the nodule bacteria showing 

peculiar structures at each end. (x 2000.) 

periblem, and dermatogen. As the yovmg root develops, a distinct protuberance 
appears on the surface of the main root, while the overlying cortical tissue of the 
latter becomes crushed by the growth of the young root through it. A few of 
these lateral roots develop normally, but the great majority become infected by 
bacteria before they emerge from the cortex, and their normal gTowth is re- 
tarded, so that they develop as the more or less spherical root-nodules. Each 
mature nodule possesses a diareh stele, as shown in Text-fig. 16, clearly de- 
limited from the cortex by an endodermis, the outer cell-walls of which are 
strongly thickened. Most of the endodermal cells contain a tannin deposit. 

The bacteria, which are clearly seen to be present when the living nodules 
are crushed out upon a slide, were isolated and pure cultures of them made. 
The surface of the nodules was sterilised in a solution composed of 2.5 c.c. of 
concentrated hydrochloric acid, 1 gram of mercuric chloride, and 500 c.c. of dis- 
tilled water, for two minutes, and subsequently washed several times in distilled 
water. The nodules were then crushed with sterile instruments upon a nutrient 



90 



STUDIES IN SYMBIOSIS, 111., 



agar medium in test tubes, so that the contents of the cortical cells were pressed 
upon the surface of the nutrient medium; the tubes were securely plugged in the 
usual manner, and incubated for several days at a temperature of 25.5° C. After 
48 hours, vigorous colonies of the bacteria were developed upon the nutrient sub- 
stratum. These colonies increased in diameter from day to day, their surface was 
raised, translucent, shining and very mucilaginous. 

Some of the bacteria were isolated upon a slide and stained in (1) gentian 
violet, Lugol's iodine, and safranin, and (2) carbol fuchsin. The bacteria are 
short rod-forms, which may occur in long chains or free. They stain deeply with 
the gentian violet; the stain is, however, almost entirely extracted with ethyl- 
alcohol (95%) and is untouched with amyl-aleohol. This reaction would suggest 
Pseudomonas radicicola according to the investigations of Harrison and Barlow 
(1906). 

The pure cultures upon nutrient agar, direct from the nodules, contained 
two kinds of individuals, namely an elliptical coccus-like form (Text-fig. 17), and 
a typical rod-shaped form. The former had a comparatively thick wall and con- 
tains one or two small granular, deeply staining bodies; it frequently occurred in 
short bead-like chains. The latter was thin-walled and at each end had a small 
circular dense area. This bacillus form frequently occurred in long chains al- 
though isolated individuals are not uncommon in cultures (Text-fig. 18). 







21 



Text-fig. 19.— A cortical cell of P. spinulosa root showing an endophytic septate 

fungus, (x 610.) 
Text-fig. 20.— A cortical cell showing the hyphae bearing chains of thick-walled 

spores, (x 610.) 
Text-fig. 21.— Peculiar forms of nuclei in cortical cells of nodule some time after 

infection, (x 900.) 

There is no doubt in my mind that these two forms are phases of the same 
organism. Firstly, each form has been demonstrated within the cortical cells of 
the nodule, the bacillus form preponderating in the young infected cells of the 
nodule, the coccus form in the older cortical cells. Secondly, when cultures were 
made from the outer, and older cortical cells only, in most cases only the coccus 
phase was present, while cultures made from the central and inner cortical cells 
contained mainly the bacillus forms. Cultures made from the whole emslied 



BY J. MCLCOKIE. 91 

nodule at first contained the eocous form in abundance and later the bacillus 
forms. The organism in the nodule of Pudocarpus spinulosa and P. elata is 
polymorphic. I cannot definitely identify it as Pseudomonas radicicola. 

The bacteria were next stvidied from the point of view of nitrogen fixation. 
For this purpose the following nutrient solution was prepared: — Cane Sugar 5 
grams, acid potassium phosphate 1 gram, Magnesium sulphate .5 gram. Calcium 
carbonate .5 gram, Distilled water 1000 c.c. Solution neutralised with Sodium- 
hydrate. 

One hundred cubic centimetres of this solution were placed in four flasks of 
300 c.c. capacity, A, B, C, D; the flasks A, B, and C were sterilised in an auto- 
clave for 30 minutes at a temperature of 140° C. and a pressure of 2| atmos- 
pheres, neutralised with NaOH and cooled. Each flask was then inoculated by 
transferring bacteria from the test-tube cultures by means of a platinum loop. 
_Th« flask D was used as a control. It was inoculated in the same manner as the 
flasks A, B and C, neutralised, then placed in the autoclave at 140" C. at 2^ atmos- 
pheres pressure for 30 minutes and cooled. The four flasks were then placed in 
an incubator at a temperature of 25.5° C. In the course of a few days the solu- 
tion in A, B and C became quite cloudy, whilst that in D remained clear. The 
flasks were incubated for 21 days at 25.5° C. and their contents were then 
examined for nitrogen by the Kjeldahl method of analysis. The following re- 
sults were obtained: — 

Nitrogen in culture solution. 

A 6 . 53 mg. 

B 7.02 „ 

C 6.79 „ 

D trace only. 

In the flasks A, B and C, a vigorous fixation of nitrogen took place, while 
■in D a mere trace was found. The obvious inference is that the increased nitrogen 
content of A, B and C was due to the living bacteria introduced, fixing the free 
atmospheric nitrogen. In the flask D, owing to the difi:erence in the treatment, 
the bacteria were killed. ISTo other conclusion can be drawn from the results of 
these experiments, than that the bacteria of the nodules of Podocarpns spinulosa 
are capable of fixing free nitrogen. 

In some of the nodules which I examined, true fungal hyphae were present 
in some of the outer cells (Text-flg. 19). These hyphae were identical in struc- 
ture with others which I observed in the cortex of the main root and forming a 
close tangle arovmd a considerable portion of the surface of the younger part of the 
root. These hyphae are generally confined to the superficial cells of the nodule 
and mother-root, but they sometimes occur in the deeper- cells of the cortex. They 
produce peculiar thick-walled, spore-like bodies in some of the cortical cells (Text- 
fig. 20, sp.). The spores arise in short chains and are formed from the hyphae 
by their division and subsequent swelling. Each cell of the spore chain has a 
thick, brownish wall, and contains one nucleus, and oil drops. These hyphae are 
probably myeorhizal in function, living symbiotically with the host root, and their 
relation to the I'oot is rather of the ectotrophie nature than endotrophie. 

Summary. 

1. The root-nodules of Podocarpns spinulosa and Podocarpns elata, like those 
of other species and genera of the Podocarpineae, are actively engaged in 
nitrogen fixation by virtue of the bacteria present in the cortical cells. 



92 STUDIES IN SYMBIOSIS^ iii., 

2. The nodules are modified lateral roots and arise from the pericycle, their 
normal growth is arrested by bacterial infection before they emerge from the 
cortex of the main root. 

5. The meristematic activity of the cells of the nodule origins is inhibited and 
their subsequent increase in size is due to the enlargement of the cells of 
the periblem. 

•i. Root hairs are commonly present upon the nodules. 

5. Bacteria enter the cortical cells of the nodule from the neighbouring cortical 
cells of the main root and multiply rapidly, spreading from cell to cell by 
means of gelatinous "zoogloea-threads" of variable width. 

(j. The zoogloeae coil in the vicinity of the nucleus, and are seldom branched. 
The active bacteria are in the form of small rods, but the inactive "bacteroids" 
are larger and oval in shape. The latter are abundant in the outer cells of 
the cortex, and in the epidennis of the nodule. 

7. In infected cells the cytoplasm becomes disorganised towards the end of the 
functional period of the nodule, the nucleus fragments, and the chromatin be- 
comes coarsely g^ranular. As many as four nuclei occur in some cells. The 
walls of the older cortical cells become strengthened by loosely arranged bars 
of thickening, giving them a traeheidal-like appearance. The zoogloeae in 
such cells gradually disintegrate with the cytoplasm and nuclei. 

8. All the outer cells of the cortex of the nodule become modified by thickening 
of their walls, and the loss of their contents. The same nodule generally 
shows all conditions between cells with active bacteria, and the empty- 
looking water-cells. 

9. The nodule may form within that of the previous season, and the outer cor- 
tical tissues therefore become crushed and distorted. 

10. The bacteria were isolated from the nodules and incubated on sterilised soUd 
and liquid media. They undoubtedly possess the capacity for the fixation of 
nitrogen. 

11. The nodules and the cortex of the main root frequently contain fungal 
hyphae and peculiar spore-like bodies belonging to the fungus. 

12. The surface of the nodule and main root is frequently invested with a loose 
tangle of fungal hyphae, some of which enter the root tissues. 

13 . The bacteria of the nodules render the free nitrogen of the atmosphere avail- 
able to the plant, and therefore make the latter independent of other forms 

of nitrogen. 

I desire to record my sincere thanks to Professor Lawson for his helpful 
suggestions. 

Literature Consulted. 

BOTTOMLEY, W. B., 1907.— The cross inoculation of the Nodule-forming Bacteria 

from Leguminous and non-leguminous Plants. Report British 

Assoc. Adv. Sc. 
— , 1912. — The Root Nodules of Myrica Gale. Ann. Bot., xxvi., Ill — 

117. 
Harrisok and Barlow^ 1900. — The Nodule Organism of the Leguminoseae. Proc. 

Roy. Soc. Canada, xii., 157. 



BY J, MCLUCKIB. 93 

McLucKiE^ J., 1922. — The Mycorhiza of Dipodium punctatum R.Br. Proc. Linn. 
Soc. N.S.W., xlvii., 293—310 ' 

McLucKiE_, J., 1922a. — The Apogeotropie Roots of Macrozamia spiralis and their 
Physiological Significance. Proc. Linn. Soc. N.S.W., xlvii., 
319—328. 

NOBBE and Hiltner^ 1899. — Die Endotrophen Mycorhiza von Podocarpus. Landr. 
Vers.-stat.y Bd., li. 

Shibata_, K., 1902. — Cytologische Studien uber die endotrophen Mycorhizen. 
Jahrh. f. Wiss. Bot., xxxvii. 

Spratt^ E. R., 1912. — The Morphology of the Root-tubercles of Ahius and Eleag- 
nus. Ann. Bot., xxvi., 120 — 128. 

, 1912a. — The Formation and Physiological Significance of Root- 
nodules in the Podoearpineae. Ann. Bat., xxvi., 801 — 814. 



94 



A EEVISION OF THE AUSTRALIAN DIPTERA BELONGING TO THE 
GENUS SARCOPHAGA. 

By Professor T. Harvey Johnston^ M.A., D.Sc, The University, Adelaide; and 

G. H. Hardy^ Walter and Eliza Hall Fellow in Economic Biology, The 

University, Brisbane. 

(Twenty-eight Text-figures.) 

[Read 30th May, 1923.] 

Australian species of Sarcophagid flies were first described in 1830 by 
Robineau Desvoidy; others were added by Macquart in 1846 and 1850, Walker in 
1849, and Thomson in 1868. In more recent times, Skuse described a species in 
1891, and Taylor another in 1917. 

During 1921-22, papers on Sarcophagid flies were published by Johnston and 
Tiegs, the specific differentiation being based upon genitalie characters and an 
attempt was made to attach to the more common species most of the thirteen 
names given by earlier authors. Parker, in 1922, described two species as new 
but, as indicated below, both these had already been described by Johnston and 
Tiegs. 

In preparing this revision it was our intention to establish a system whereby 
identifications could be made from female specimens and our study of the chaeto- 
taxy was undertaken mainly for this purpose, but it was soon found that a 
critical study of the genital parts had to be undertaken first so as to eliminate 
errors due to imperfections of previously studied material and to the misunder- 
standing of the limits of variation in the male. We are convinced that the 
chaetotaxy of the female conforms to that of the male in regard to certain 
features which may be used in specific determination, and in several cases, where 
this rule has not V^een applicable, it is possible that an error has been made by 
attaching wrong identification labels to material previously studied. 

This revision completes the first step towards establishing a satisfactory 
taxonomic treatment of Australian Sarcophagid flies in so far as it revises the 
treatment of male genital parts, giving descriptions as well as drawings of out- 
standing features and adding a comparative study of the chaetotaxy, that por- 
tion relating to the abdomen and femora not having been attempted before. 

The new names added to tlio Australian list include those of one new species, 
one previously wrongly identified, for which a new name is proposed below, and 
also for one that previously was only recorded from Europe and North America, 
namely S. secunfera Villcn. Eight names are placed as synonyms for the first 
time. 



BY T. HARVEY JOHNSTON AND G. H. HARDY. 95 

The term "Australian," as used in this paper, does not include localities such 
as Lord Howe Island, New Guinea and islands in its vicinity. 

The first key for the determination of males is based entirely upon genital 
characters. The second key is based mainly on a selection from the chaetotaxy 
and other characters that we consider sufficiently constant to warrant their use in 
this way. The third key for the determination of the females is the first pub- 
lished attempt to isolate females, mainly on the chaetotaxy, and can include, of 
course, only species of which the females are known. Until more material is 
bred and the females determined in this manner, it is uncertain whether our key 
will ultimately be found reliable, though we believe it will prove to be so in 
most cases. 

Acknowledgments. — We are indebted to Dr. C. Anderson and Mr. H. A. 
LongTuan, the Directors of the Australian and Queensland Museums respectively, 
and to Mr. G. F. Hill, Entomologist of the Institute of Tropical Medicine, Towns- 
ville, for the loan of those types of Sarcophagid flies that are respectively under 
their charge; to Dr. E. W. Ferguson, Mr. G. H. Dutton and others for the loan 
of specimens in their collections. 

From Mr. R. R. Parker we received a letter in which he informed us that 
his two species, described as new, were forestalled by Johnston and Tieg-s, and 
suggested certain further synonymy to which we have given due consideration. 

Chaetotaxy. 

(A). The bristles of the head (Text-fig. 1) are classified as follows: — 

Vertical: Two pairs, inner and outer, situated in a line behind the ocelli; the 
inner pair always long and convergent, the outer pair always divergent, shorter 
and situated near the corner of the eyes and, in the male only, may be reduced 
in size or even obsolete. 

Frontal: On each side of the frontal stripe, a row of bristles descends from 
below the oeeUi to the base of the antennae, whence the rows usually diverge for 
a short distance. 

Fronto-orbital: Below" the outer vertical and above the frontal, near the eje- 
margin, there is a reclinate bristle. On the female only there are, besides this, 
two further bristles placed lower; these form a line of three bristles parallel to 
the eye-margin. 

Vibrissa: A pair of long, stout bristles situated near the oral. margin and 
excessively long in comparison with the other bristles around them. 

Facial: A series of bristles, small and often hair-like, extending upwards 
from each vibrissa. A second, or part of a second, row is present but only the 
number of bristles in the longer row is quoted herein. 

Oral: A series of bristles passing from each vibrissa along the oral margin 
and often contrasting with the hairs around them but sometimes not so easily 
detected. We have followed the policy of counting these bristles only when they 
are black. 

Post-ocular: Immediately behind, and parallel to, the eyes there is a row of 
black bristles often referred to as post-orbital. Sometimes the hairs behind this 
series form one or two parallel rows of black bristles, these extra rows being ap- 
parently constant when they occur on a species and forming one of the leading 
clues to the identity of females. 

Post-vertical: Two pairs of bristles situated behind the verticals. 



96 



AUSTBALIAN DIPTERA BELONGING TO THE GENUS SAROOPHAGA, 



(B.)- The terms used for the bristles on the thorax are as follows: — 
Humeral: Three bristles inserted on the humeral callus. The unique S. 
bancrofti contains a fourth, but this character is paralleled by a bristle-like hair 
occasionally seen in isolated cases on other species and, therefore, we do not con- 
sider S. bancrofti to be normal in this respect. 

Post-humeral: Two bristles situated on the area immediately adjacent to the 
humeral callus. 




... HP.. ^^ 
Scut. 

Text-fig. 1. Diagram of the head. 
Fac. facial; F-o. frontal orbital; Fr. frontal; iV. inner vertical; Oc. ocellus 
on which the ocellar bristles occur; Or. oral; oV. outer vertical; Po. postocular; 
Pv. postvertical ; Vih. vibrissa. 

Text-fig. 2. Diagram of the thorax. 
Ac. acrostichal; Dc. dorsocentral ; Hp. hypopleural; Hum. humeral; I-a. 
intra-alar; Mp. mesopleural; Np. notopleural; Pa. post-alar; Ph. post-humeral; 
Pp. propleural; Ps. presutural; Ptp. pteropleural ; S-a. supra-alar; Scut, seutellar; 
Sp. sternopleural ; t.s. transverse suture. 



Notopleural: Fou- bristles alternately short and long, situated immediately 
above the dorsopleural suture, Ijctween the humeral callus and the base of the 
wing. 

Presutural: A rather strong bristle situated immediately before the trans- 



BY T. HAEVEY JOHNSTON AND G. H. HARDY. 97 

verse suture and above the notopleural bristles. A little beyond this may be an- 
other smaller bristle, in a line with the intra-alars, but we prefer to consider thig 
second bristle as one of the intra-alars rather than a second presutural. 

Supra-alar: Immediately above the base of the wing are four bristles, the 
fourth sometimes small or even minute, but invariably indicated. 

Intra-alar : Two or more bristles parallel to the supra-alars, but further from 
the base of the wing, appear to be of some specific value. A third bristle is often 
present, and even a fourth may occur, but situated anteriorly to the transverse 
suture (see presutural). 

Post-alar: Two bristles situated on the post-alar callus. 

Dorsocentral: Reaching from just behind the head to the scutellum there is 
a complete row of bristles, of which four at least are situated between the trans- 
verse suture and the scutellum in the genus Sarcopliaga, and only three in the 
genus Helicohia. 

Acrostichal : Between the two dorsocentral rows, vestiges of two further rows 
may be traced. These brifetles, when present, just anterior to the transverse 
suture, are termed presutural, and those just anterior to the scutellum are pre- 
seutellar. The presence or absence of bristles in these particular positions is of 
considerable specific importance. 

Scutellar: On the scutellum there are four pairs of bristles in the male and 
three in the female. 

Propleural: Two bristles situated immediately above the anterior coxa. 

Mesopleural: A row of five bristles along the mesopleural suture. 

Sternopleural: Three bristles arranged in a row (1.1.1.) on the sclerite an- 
terior and adjacent to the intermediate coxa. Also between the anterior and 
intermediate legs a row similar in appearance to those on the anterior and inter- 
mediate coxa. 

Pteropleural: A group of about three bristles just below the wing. 

Hypopleural: A vertical row of slender bristle? just above the hind coxa. 

(C). The bristles on the abdomen are as follows: — 

On the first segment there are discal bristles situated on each side, varying 
in number from one to about eight or more and not particularly constant on any 
one species, but a more or less useful character is to be found in the manner in 
which they are arranged. They form one or two distinct rows, each composed of 
one or several bristles. When they are considerably reduced in number, one row 
may be eliminated on a species that normally has two rows; this reduction is 
seldom found on both sides of the insect, so that when a second row is not de- 
tected on one side it may be indicated on the other. Besides these discal bristles 
there are from one to three siibmarginal bristles placed near them. 

The second segment is generally provided with from one to three lateral siih- 
marginals. In S. howensis (from Lord Howe Island) alone, of all species of 
Sarcophaga known to us, there is also a median pair of submarginal bristles 
strongly developed and equivalent to those occurring on the third segment of this 
and all other species of the genus. The lateral submarginals may be small or even 
obsolete on the female. 

The third segment has a median pair of submarginal bristles, and from one 
to three laterals on each side. These median and lateral submarginals mav ^ be 
joined by intermediate bristles, making up to six pairs of submarginals similar 
to those on the fourth segment. 



98 AUSTRALIAN DIPTERA BELONGING TO THE GENUS SAROOPIIAGA, 

The fourth segment has five, or more usually six, pairs of submarginals on 
the dorsal side. Except between the median pair, these alternate with marginal 
bristles that may be small and hair-like or may be strongly developed, although 
not so stout as the submarginals. These marginals continue ventrally, where 
they are densely massed and often excessively long and slender. 

(D). The bristles on the legs are referred to as follows: — 

The femur is more or less oval in cross-section and that flattened surface 
that faces the head when the legs are at right angles to the body is called the 
anterior side. The other three sides are respectively dorsal, posterior and ventral. 

The anterior femur is without bristles on the anterior side and has a row on 
each of the other three. 

The intermediate femur normally has one anterior row, one posterior row 
which is restricted to one, two or three subapical bristles, and two ventral rows 
between Avhich, when reflexed, the tibia lies. 

The posterior femur has bristles similar to those on the intermediate femur 
with the addition of a dorsal row. 

The bristles on the female femora are generally fewer in number within 
each row than those on the male; otherwise the variations from the normal that 
occur in the male will usually be found occurring in the female. It should be 
noted that where a row on the male is represented by onlj^ one or two bristles, 
in the female it may be obsolete. 

The relative lengths of pubescence, short hair and long hair are illustrated 
in Text-fig. 4. ,-: ^ "f'^'f, 

Method of pinning specimens and setting the male genitalia. 

The method we have adopted in setting Sarcophagid flies is to use a small 
fine pin passed through the thorax behind the transverse suture and well in front 
of the scutellum. In this manner the acrostiehal bristles remain undamaged and 
are readily perceptible if present. The insect, if a female, is then "staged." 

The small pin referred to plays an important part when extracting the male 
genitalia. It allows the insect to be placed sideways with the genital segments 
lying close to the cork to which the insect is pinned. A second pin is placed in 
the cork just above the tip of the abdomen so as to give rigidity and then the 
genital organs are exposed by inserting the point of another pin against the 
forceps and drawing them outwards. With fresh material the penis will readily 
fall into position but with some relaxed specimens a further pin will be necessary 
to keep the organ in place. 

To bring about relaxation we dip the tip of the abdomen into distilled water 
heated just below boiling point. We find that although the wings may be slightly 
damaged, especially near the tip, the insect retains its colour and form far more 
satisfactorily than is usually the case with those relaxed slowly by the more 
general methods. The time taken to relax by our method is usually half a 
minute but depends upon the age of the specimen, some very old ones taking 
quite a minute. 

Method of examivirtg and drawing the genital parts. 

In examining the type , speci mens and drawing the genital parts, we have 
utilised a Leitz-Greenoucrh Sterenscnpic Binocular Microscope with eye-pieces 0-V, 
and objectives 25 and 48 mm. The parts of the penis are so very small that it 



BY T. HARVEY JOPINSTON AND G. H. HARDY. 



99 



is necessary to have a stereoscopic effect in order to understand them. We have 
found it advantageous to examine the organ under higher powers in the direct 
sunlight, at the same time reflecting the rays with a concave miiTor on to the 
object thus illuminating the inside as well as the outside of the genital complex. 




Cost. 



Text-fig. 3. Diagram of the abdomen. 
I.-IV. 1st to 4th segments; disc, diseal; sm. submarginal; m. marginal. 
The median pair of submarginal bristles on the third segment have been 
accidentally omitted in the drawing. 

Text-fig. 4. Diagram of the posterior femur and tibia. 

l.h. long hair; pub. pubescence; s.h. short hair. 

Text-fig. 5. Diagram of the wing. 

The veins are named in rotation as follows. — Cost, the costa which, bordering 
the wing, reaches to near the apex; hm. humeral, a small cross vein; Sc. sub- 
costa; Ri. first radial; R2+3. second and third radials united; R4+5. fourth and 
fifth radials united; Mi. first median; M2. second median; Ms and M4. third and 
fourth median veins respectively; the basal portion is obsolete thus leaving these 
two remnants acting as a cross vein between the first median and the cubital 
vein; Cui first cubital which unites with M4 at its apex; An. anal; m.c. median 
cell. 



By this method we have been able to trace the actual outline of each part 
of the penis to its source and in a few cases we bave augmented our studies by 



lOO AUSTRALIAN DIPTEEA BELONGING TO THE GENUS SARCOPHAGA^ 

mieroslides made from specimens other than the types. The genital parts vere 
drawn in outline by the aid of a camera lueida and the lines not visible by this 
method were drawn in freehand. 

In order to show the relative shape of the various parts more than one view 
of the penis has been drawn. In each ease the usual lateral view is given and 
another aspect such as the ventral or anterior side supplements it. A ventral 
view, correctly given, would sometimes hide an important aspect of the organ, so 
in such cases a drawing is made from a more or less tilted position. 

Explanation of terms used when describing the male genitalia. 

In studying the genital parts we have found a set of characters that occur 
in the majority of species and are constant in outline and design within each 
species. In the case of the penis we have had to select terms in order that these 
parts may be described. 

Forceps : A pair of appendages that protect the genital opening when the 
penis is withdrawn. 

Claspers : On each side of the penis there is a pair of pointed and f orwardly 
directed processes, the first of which is called the anterior, the hinder the posterior 
clasper. The anterior elasper is movable in a longitudinal, and the posterior in 
a lateral direction. 

First joint of the penis : The basal joint. 

Second joint of the penis: The apical joint, which is very complex, contains 
the following parts: — 

Anterior appendage: Often large and consisting of one, sometimes two, pairs 
of flanges that vary greatly in size and shape in the various species. 

Lobe: On each side of the main and enveloping portion of the penis is a 
flange that is often lobe-like and sometimes forms a conspicuous process. 

Lateral process : Situated beyond the lobe is a pair of forwardly directed pro- 
cesses, always conspicuously separated at the base and never contiguous. 

Apical process: An extension of the apex, if present, is referred to by this 
term. It may consist of an unpaired process or a pair of closely adjacent parts. 
Extra processes may also be present, but are not usual. 

Interior process : A term given to a process arising internally from just pos- 
terior to each lobe ; so'f ar this process has only been detected in S. peregrina Desv. 
and S. kappa J. and T. 

Filaments: Between the anterior appendage and the main enveloping portion 
of the penis there may exist a further pair of appendages referred to by this 
name. They can be seen arising from the base of the joint and are generally 
filamentous, although in one species, S. securifera Villen., they appear to be 
highly modified at the apex. Filaments are not always observable and as far as 
we have determined do not exist in species where they are not apparent. 

Variation in the size of speciea. 

One of tlie commonest species of Sarcophagid flies around Brisbane is S. 
tryoni and of this we have bred over one hundred and fifty specimens, from at 
least ten females, during four months. One captured female, a very large speci- 
men, produced a batch of fiios, all of which were of a large size, about 12 mm. 
in length. Another, a small female, the identity of which was not known at the 
time of capture, produced a batch of flies ranging from 6 mm. to about 9 mm. 



f BY T. HARVEY JOHNSTON AND G. H. HARDY. 101 

These two batches, as far as we could tell, were bred under similar conditions and 
were bred out within a week of each other, consequently it seems possible that 
size may be an hereditary character quite apart from food. We have other 
species, varying in size, but not quite reaching the extremes found in ;S'. tryoni; 
nevertheless there is sufficient to show that dimensions are too unstable a charac- 
ter for specific determination. S. omega and S. eta range from 7 to 12 mm. in 
the various collections examined. 

It is significant to note that the chaetotaxy varied remarkably little in the 
smaller and larger batches of S. tryoni that were bred, and it agrees closely with 
that of captured specimens. 

Genus Sarcophaga Meigen. 

Definition. — The arista is strongly plumose on the basal half and bare 
thence to the apex. Eyes bare. Thorax with three distinct black stripes. The 
second humeral bristle situated in a line with or above the presutural. At least 
four post-sutural dorsocentral bristles. Abdomen tessellated with silvery and 
black tomentum. Wings without bristles on the vein Ri, and with a few at the 
base of R4-5. The vein Mi, at half its length, bends sharply upwards whence, 
for a short distance, it proceeds at right angles to the basal part, thence, diverg- 
ing, reaches the costa before the apex of the wing and near the apex of the vein 
R4-5. The venation is illustrated and explained in Text-fig. 5. 

Observations. — The above definition will be sufficient to distinguish species of 
the genus Sarcophaga, as recognised by us, from other genera of Sarcophagid 
flies known to us and also from the numerous flies which, although belonging to 
distinct groups, have a very striking resemblance to these. Further characters 
have been given under the discussion on the chaetotaxy. 

Under the above definition will come also many of the species that have 
been made the types of other genera, thirty-nine of which were diagnosed in key 
form by Townsend in 1917. We have attempted to place our Australian species 
into groups in accordance with Townsend's key, but have found that, in such an 
attempt, species that, according to genitalia, must be considered closely allied, 
were widely separated. 

In more recent literature several genera have been proposed to contain 
groups of species that have closely allied genital characters and it is on this 
principle that ultimate success in re-grouping the Sarcophagid flies will probably 
be attained. 

Sarcophaga alpha, S. Beta, S. beta and S. howensis, to which also may be 
added (at least for the time being) S. tryoni, S. impatiens and even perhaps S. 
epsilon, form one group. All these have a well developed flange on the anterior 
clasper which in the case of the first four becomes bifid. All except the last- 
named have also a very strongly developed pair of forceps. 

S. misera, S. kohla and perhaps aurifrons, omega and eta will form a second 
group. Each has a pair of well developed lateral processes which are bifid in 
the first two species. 

S. bancrofti and S. fergusoni will form yet another group, whilst the re- 
maining species have their genitalia too diverse in structure to permit of their 
being readily grouped in this manner. 

A study of the nearest allies within these groups shows that an attempt to 
use chaetotaxy as a guide to affinities would be misleading. 



102 AUSTRALIAN DIPTERA BELOKGING TO THE G:ENUS SARCOPHAGA, 

Key to the species of Australian flies of the genus Sarcophaga hased upon th6 

male genital characters. 

1 . Genital segments red securifera Villen. 

Genital segments black 2. 

2. Anterior clasper bifid, or with a very wide flange 3. 

Anterior clasper simple, or if a flange be present then it does 

not exceed the width of the clasper 9. 

3. Anterior clasper bifid and with a knob anterior to it 4. 

Anterior clasper not bifid and without a knob anterior to it. . . 6. 

4. Filaments present 5. 

Filaments absent ;seta J. & T. 

5. Anterior clasper bifid nearly to its base alpha J. & T, 

Anterior clasper bifid for only half its length beta J. & T. 

6. Filaments present 7. 

Filaments absent tryoni J. & T. 

7. Forceps large; seen laterally curved and with a conspicuous angle at two- 

thirds its length. Apex of the anterior appendage with a conspicuous 

forvvardly directed "hook." impatiens Walk. 

Forceps of moderate length and not angiilated. Anterior ap- 
pendage without a hook 8. 

8. Anterior appendage and lobe very small. Lateral and apical processes to- 

gether forming a complex tapering and forwardly directed termination to 

the penis epsilon J. & T. 

Anterior appendage large and expanding. Lobe produced to a pointed apex. 
Lateral process small and curved under a wide, lateral flange on the 
apical process hardyi J. & T. 

9. Lateral process present ..10. 

Lateral process absent 18. 

10. A pair of knobs situated behind the lateral processes which are very 
complex omikron J. & T. 

Without such knobs 11. 

n. Lateral process bifid towards the tip 12. 

Lateral process simple 13. 

12. Anterior appendage composed of one pair of broad flanges. . . misera Walk. 
Anterior appendage composed of two pairs of narrow flanges. . . kohla, n.sp. 

13. With an apical process 14. 

Without an apical process 16. 

14. Apex of the anterior appendage produced into an extremely long process 

directed downwards eta J. c^: T. 

Anterior appendage without such a process .15. 

15. Anterior appendage broad and much of its area consisting of a mass of 

minute spines peregrina R.D. 

Anterior appendage slender and scarcely di.scernible when the penis is seen 
laterally depressa R.D. 

16. Filaments present aurifrons Maeq. 

Filaments absent 17. 

17. Lateral process long, lobe long and pointed; anterior clasper conspicuously 

longer than the posterior omega J. & T. 

Lateral process short, lobe inconspicuous; claspers of about equal length. .. 

froggatti Taylor. 

18. Filaments present 19. 

Filaments absent 20. 

19. Apex of the forceps minutely bifid. Anterior clasper with a small flange at 

the extreme apex. The pair of anterior appendages slender and bowed, 
almost meeting at the apex synia, n.sp. 



BY T. IIARVEY JOHNSTON AND G. H. HARDY. 103 

Apex of the forceps simple. Anterior clasper much longer than the posterior 
and without a flange at the apex. Anterior appendage widely expanding 
and broad gamma J. & T. 

20. Apical process paired (i^e. consisting of two contiguous pro- 

cesses) 21. 

Apical process not paired fergusoni J. & T. 

21. Anterior appendage doubled back, so that the apex is directed towards the 

base hancrofti J. & T. 

Anterior appendage directed downwards in the usual manner. . . 22. 

22. Claspers of about equal length. A pair of interior appendages protrude 

slightly beyond the apical appendages kappa J. & T. 

Anterior clasper longer than the posterior. Without interior appendages. . . 

littoralis J. & T. 

Alternative key to the male flies of the genus Sarcophaga. 

1 . Genital segments red securifera Vilien. 

Genital segments black 2. 

2. With one row of bristles behind the eyes 3. 

With two rows of bristles behind the eyes 16. 

With three rows of bristles behind the eyes 19. 

3. With long hair on the posterior tibiae 4. 

Without long hair onthe posterior tibiae 13. 

4. With prescutellar acrostichal bristles 5. 

Without prescutellar acrostichal bristles; with presutural acrostichals 

. tryoni J. & T. 

5. With two rows of ventral bristles on posterior femur 6. 

With one row of ventral bristles on posterior femur 10. 

6. Long hair on posterior tibia abtmdant 7. 

Long hair on posterior tibia scanty zeta J. <% T. 

7. With one row of disco-lateral bristles on first abdominal seg- 

ment 8. 

With two rows of disco-lateral bristles on first abdominal segment 

impatiens Walk. 

8. Species with golden j-ellow tomentum on head and thorax 9. 

Species with silvery-grey tomentum on head and thorax . . . . hardyi J. & T. 

9. Third joint of the antennae about three times the length of the second. .. 

alpha J. & T. 

Third joint of the antennae about twice the length of the second, beta J. & T. 

10. The marginal bristles on the fourth abdominal segment strongly developed, 

almost as long as the submarginals 11. 

The marginal bristles on the fourth abdominal segm.ent weak or obsolete. 

kappa J. & T. 

11. Third antennal joint about three timies the length of the second 

omikron J. & T. 

Third antennal joint about twice the length of the second. . . 12. 

12. Outer vertical bristles absent, or at most as long as the post-oculars 

omega J. & T. 

Outer vertical bristles twice the length of the post-oculars. . . gamma J. & T. 

13. With prescutellar acrostichals 14. 

Without prescutellar acrostichals froggatti Taj'lor 

14. With presutural acrostichals 15 

Without presutural acrostichals misera Walk. 

15. With two extremely long bristles at about half the length of second ventral 

row on intermediate femur. "\^entral hairs on abdomen unusually long. 

depressa R.D. 



104 AUSTRALIAN DIPTERA BELONGING TO THE GENUS SARCOPHAGA^ 

Bristles of second ventral row on intermediate femur of about equal length. 

littoralis J. & T. 

16. With long hair on the posterior tibiae 17. 

Without long hair on the posterior tibiae hanerofti J. & T. 

17. With two rows of ventral bristles on intermediate and pos- 

terior femora 18. 

With one row of ventral bristles on intermediate and posterior femora 

eta J. & T. 

18. Ventral hair of abdomen short and scanty ', hohla, n.sp. 

Ventral hair of abdomen rather long and dense fergttsoni J. -.% T. 

19. With prescutellar acrostichals; without long hairs on posterior 

tibiae 20. 

Without prescutellar acrostichals; with long hairs on posterior tibiae 

epsilon J. & T. 

20. With one row of ventral bristles on intermediate and posterior 

femora 21. 

With two rows of ventral bristles on intermediate and posterior femora 

aurifrons Maeq. 

21. With pre sutural acrostichals synia, n.sp. 

Without presutural acrostichals peregrina R.I). 

Key to the female flies of the genus Sarcophaga. 

1. Genital segments red securifera 

Genital segments black or brown 2. 

2. With one row of postocular bristles 3. 

With two (or more) rows of postoculars 8. 

3. With presutural acrostichal bristles 4. 

Without prestitural acrostichals; with golden yellow tomentum behind the 

eyes misera. 

4. With prescutellar acrostichals 5. 

Without prescutellar acrostichals 7. 

5. With one row of disco-lateral bristles on the first abdominal segment. 6. 
With two rows of disco-laterals on first abdominal segment; with golden 

yellow tomentum behind the eyes impatiens, kappa, omega. 

8. With golden yellow tomentum behind the eyes. . . . beta, depressa, omikron. 
With silver grey tomentum behind the eyes hardyi, littoralis. 

7. With three postsutural intra-alar bristles truoni.* 

With two postsutural intra-alars . . . . froggatti. 

S. With presutural acrostichals; with silver grey tomentum behind the eyes. .. 

peregrina. 

Without presutural acrostichals 9. 

9. With golden yellow tomentum behind the eyes aurifrons. 

With silver-grey tomentum behind the eyes eta. 

In choo.sing a set of characters whereby any female specimen can be relegated 
to its correct species, we have been confronted with several difficulties. The 
females of many species are not certainly known to us, and in only a few cases 
have we a sufificiently large number that have been correctly referred to their re- 
spective males either by breeding or by having been taken in copula. Moreover, 
there appear to be several cases where the identification label has been trans- 
ferred on specimens previously studied, thus causing confusion in I'egarcl to the 
limits of variation within the species. 

*Since this key was prepared further specimens of S. tryoni have been bred 
and many of them are without the third intra-alar bristle. 



BY T, HARVEY JOHNSTON AND G. H. HAKDY. 105 

Of the twenty-two species recorded from Australia, the females of seven are 
unknown; seven are isolated in the above key, and eight are further grouped into 
three groups not containing more than three species each. We believe a little 
modification of the scheme here outlined, when amplified by other characters that 
may be discovered by a critical survey of females, will ultimately lead to the 
compilation of a key that will permit the ready determination of the species from 
the female sex. 

Sarcophaga alpha Johnston and Tiegs. (Text-fig. 6.) 

Sarcophaga alpha, Johnston and Tiegs, Proc. Roy. Soe. Queensland, xxxiii., 
1921, p. 57, fig. 21 ; Rec. Austr. Mus., xiii., 1922, p. 177. 

Synonymy. — The illustration given by Bottcher, when describing his S. anti- 
ope (a species from Formosa and New Guinea), has much in common with S. 
alpha and we consider it possible that the two may ultimately be found to belong 
to one species. The penis in Bottcher's figure does not quite conform to our il- 
lustration although the essential characters are there. 

Description. — d*. Head. Outer vertical bristles about twice the length of 
the post-oculars; twelve frontals; two or three facials; nine orals; one row of 
post-oculars. 

Thorax. Two intra-alar ; one small pair of presutural, and two large 
pairs of prescutellar acrostichals. 

Abdomen. On the first segment one row of two diseal lateral bristles and 
one submarginal; an extra submarginal is developed on one side in the holotype. 
On the second, one lateral submarginal and an extra one on one side in the holo- 
type. On the third, one median pair and three lateral submarginals. On the 
fourth segment six pairs of submarginals alternating with marginals. 

Genitalia. Forceps long and strongly curved. The anterior clasper 
bifid almost to the base, and anterior to this there is a small knob; the posterior 
clasper is broader than the anterior but about the same length. Anterior ap- 
pendage rather broad and simple; apical process tapering, but moderately broad 
and rounded at the apex; filaments very long. 

Legs. The ehaetotaxy conforms to the general type ; long hair is abundant 
on all the femora and on posterior tibiae. 

Hah. — Queensland. 

Observations. — The above description is taken from the holotype specimen in 
the Queensland Museum. By slightly altering the position of the parts and al- 
lowing for the more forwardly placed filaments, an outline of the genitalia very 
similar to that given by Bottcher from his S. antiope can be seen on the holo- 
type of S. alpha. 

Sarcophaga zeta Johnston and Tiegs. (Text-fig. 7.) 

Sarcophaga zeta, Johnston and Tiegs, Proc. Roy. Soe. Queensland, xxxiii., 
1921, p. 76, fig. 20; Also xxxiv., 1922, p. 183; Rec. Austr. Mus., xiii., 1922,- p. 180. 

Description. — c?. H e a d. Outer vertical bristles inconspicuous, as long as 
the post-oculars; thirteen frontals; eight or more facials; about five orals; one 
row of post-oculars 

Thorax. Two intra-alar bristles ; presutural and prescutellar acrostichals 
present. 



106 



AUSTRALIAN DIPTERA BELONGING TO THE GENUS SARCOPHAGA, 



Abdomen. On the first segment a row of two diseal lateral bristles and 
one submarginal. On the second, one lateral submarginal. On the third, one 
median pair and two lateral submarginals. On the fourth segment the bristles 
conform to those on S. alpha, but this cannot be detected on the holotype as the 
segment is broken. 




Texfc-fig. 6. Sarcophaga alpha J. & T. 
Text-fig. 7. Sarcophaga zeta J. & T. 
Text-ng. 8. Sarcophaga beta J. & T. 



The following letters occurring in the above and all subsequent illustrations 
indicate the names of the various parts : — 

A. Forceps or the apical part of them seen posteriorly; B. The same, lateral 
view; C. Clampers; D. Another aspect of one or both the claspers; E. Second 
joint of the penis seen laterally; F. The sam.e, ventral view; G. The same, or its 
apex seen posteriorly; H. The same, seen anteriorly; I. Second joint of the 
penis, or part of it drawn from another specimen belonging to the same species. 

Claspers. a.c. anterior clasper; 'p.c. posterior clasper; k. knob. 

Penis, a.a. anterior appendage; a.j). apical process; p,. filaments, {.p. interior 
process; /c. knob; I. lobe; Lp. lateral process. 



BY T. HARVEY JOHNSTON AND G. H. HARDY. 107 

Genitalia. Forceps and claspers identical with those of S. alpha. An- 
terior appendage enfolded by the remainder of the second joint of the penis 
which has a pair of slender apical processes; filaments absent. 

Legs. The chaetotaxy conforms to the general type but the second ventral 
row on the posterior femur is weak. The femora are scantily clothed with long 
hail", the majority being found on the anterior femur; hairs on the posterior 
tibiae are also scanty. 

Hah. — Queensland. 

Observations. — The above description is based upon the holotype specimen 
in the Queensland Museum. In order to test the structure of the penis we re- 
laxed a paratype specimen and could find no sign of the filaments. At the same 
time we found that, when relaxed, the posterior clasper was movable in a lateral 
direction and the anterior clasper was hinged to move backwards^ so that the 
three apices of the claspers were brought into one line, in which position they 
were allowed to dry on one side. The position was secured again on a fresh 
specimen and with remarkable ease, the two claspers falling into place as readily 
as they had done in the relaxed specimen. 

Saroophaga beta Johnston and Tieg*s. (Text-fig. 8.) 

Sarcophaga beta, Johnston and Tiegs, Proc. Roy. Soe. Queensland, xxxiii., 
1921, p. 58, fig-. 6; Rec. Austr. Mus., xiii., 1922, p. 177.— S. delta, Johnston and 
Tiegs, ibidem, xxxiii., 1921, p. 62, fig. 13. 

Synonymy. — Sarcophaga delta was described from a single male and the penis 
is doubled back, somewhat distorting the parts, but we have given this a more 
critical examination, as the other parts of the genitalia were found to be identical 
with those of ^S*. beta. As the parts of the penis are essentially the same in the 
holotypes of these two species we consider that they cannot be maintained as 
separate species. 

Description. — S. Head. Outer verticals about as long as the post-oculars; 
on the holotype of 8. beta there are nine frontals on one side and twelve on the 
other; and on S. delta ten on each side; facials seven or eight; orals seven, and 
on 8. delta this row is doubled, that is, there are two rows of about seven each; 
one row of post-oculars. 

Thorax. Intra-alar bristles two {S. beta) or three {S. delta). Owing to 
damage, on the thorax only prescutellar acrostiehals are discernible on the holo- 
types, but presutural acrostiehals are also present on other specimens. 

Abdomen. On the first segment one row of four {S. beta) or five {S. 
delta) discal laterals, and one submarginal. On the second, two lateral sub- 
marginals. On the third, one median pair and two lateral submarginals. On the 
fourth segment five pairs of submarginals. 

Genitalia. The forceps, claspers and the knob are not very difEerent 
from those on S. alpha, but the anterior clasper is bifid for only half its length. 
The anterior appendage contains a very delicate fringe that is not easily detected, 
owing to the lack of chitinisation on the fringed parts; apical process rather 
broad; filaments very long. These genital parts are very similar to those of S. 
howensis Johnston and Hardy, from which they differ only in the apex of the 
second joint of the penis. 

Legs. The chaetotaxy eonfoi-ms to the general type, but with the addition 
of two subapieal bristles representing a second dorsal row on the posterior femur. 
Long hair occurs on all femora and on posterior tibiae. 



108 AUSTRALIAN DIPTERA BELONGING TO THE GENUS SARCOPHAGA^ 

^. The chaetotaxy of the female corresponds remarkably well after allow- 
ance is made for sexual differences. There are twelve frontal bristles on each 
side. 

Hob . — Queensland. 

Observations. — The above description is taken from the holotype specimens 
of S. beta and S. delta, and the allotype of iS*. beta, all of which are in the 
Queensland Museum. A further series of specimens has been examined by us 
including eight males; the variations in chaetotaxy cover those of the two hole- 
types, but we have not been able to detect within the series corresponding dif- 
ferences in the male genitalia. 

The species ali^ha, zeta, howensis and beta have many genital characters in 
common; the bifid anterior clasper with the curious knob anterior to it, the long 
forceps conspicuously bent at two-thirds their length, the small lobe, and the 
absence of the lateral process do not exhaust the points of similarity. The for- 
ceps are raised upon a tubercle that is very prominent in comparison with the 
corresponding structure on other species. This combination of characters ap- 
pears to point to very close affinities and it is significant to note that one species, 
S. hoivensis, has the abnormal character of a pair of median submarginal bristles 
on the second abdominal seginent, which indicates that outstanding characters in 
chaetotaxy cannot be used advantageously to form natural groups within the 
genus Sarcophaga. 

Sarcophaga tryoni Johnston and Tiegs. (Text-fig. 9.) 

Sarcophaga tryoni, Johnston and Tiegs, Proe. Roy. Soc. Queensland, xxxiii., 
1921, p. 54, figs. 9-10; Rec. Austr. Mus., xiii., 1922, p. 176. — S. queenslandae, 
Parker, Canadian Entom., liv., 1922, p. 6, figs. 1-2. 

Synonymy. — Mr. R. R. Parker informs us that his species S. queenslandae 
is the same as S. tryoni. 

Description. — cj". Head. Outer verticals slightly longer than post-oculars; 
twelve frontals on one side, thirteen on the other; one strong and several small 
facials; about seven orals; one row of post-oculars. 

Thorax. Two intra-alars on one side, three on the other, and a further 
one placed anteriorly to the suture is also present. Only the presutural aeros- 
tichals present. 

Abdomen. On the first segment one row of four or more discal lateral 
bristles and one submarginal. On the second, one lateral submarginal. On the 
third, one median pair and two lateral submarginals. On the fourth segment six 
pairs of submarginals. 

Genitalia. Forceps long. Anterior clasper with a very broad flange. 
Anterior appendage formed of two parts, the lower of which contains a small 
but distinct process on the ventral edge; lobe pointed; lateral process barely in- 
dicated by a small pointed projection just above the lobe; a pair of apical pro- 
cesses, each containing three pointed projections. 

Legs. The eliaetotaxy conforms to the general type, but the second ventral 
row on the posterior femur is I'epresented by three or four subapieal bristles 
only. Long hair occurs scantily on the anterior and posterior femora and abun- 
dantly on the posterior tibiae. 

?. The chaetotaxy of the female corresponds well with that of the male, 
after allowance is made for sexual differences; the submarginal lateral bristle on 



BY T. HARVEY JOHNSTON AND G. H. HARDY, 



109 



the first abdominal segment is missing and the second ventral row on the pos- 
terior femur is well indicated. 

Ha b . — Queensland. 

Observations. — On re-examining the material in the Queensland Museum we 
discovered that the holotype label had been attached to a female specimen, so 
we have transferred this to the specimen from which the "figure 9" of the 
original drawings was made. The above description is taken from this newly 
selected holotype and the allotype, both of which are in the Queensland Museum. 
Over two hundred specimens have been examined by us, and it appears that the 




! i V^-'f / j: ' 

1 1 ^■p-\e^tdj''''- 



Text-iig. 9. Sarcophaga tryoni J. & T. 
Text-ti?. 10. Sarcophaga impatiens Walker. 
Text-fig. 11. Sarcophaga epsilon J. & T. 

absence of prescutellar bristles, together with one row of post-oculars and three 
post-sutural intra-alar bristles are characters ample for the determination of this 
species from either sex. 



Sarcophaga impatiens Walker. (Text-fig. 10.) 

Sarcophaga impatiens, Walker, List Dipt. Brit. Mus., iv., 1849, p. 828; 
Johnston and Tiegs, Proe. Roy. Soc. Queensland, xxxiii., 1921, p. 52, figs. 18-19; 
Also xxxiv., 1922, p. 182; Ree. Austr. Mus.. xiii., 1922, p. 176. 



110 AUSTRALIAN DIPTERA BELONGING TO THE GENUS SARCOPHAGA, 

Status. — A female specimen was compared by Major E. E. Austen with the 
holotype female in the British Museum. Bred specimens were determined as 
belonging to the same species as that compared with the type and the male was 
described for the first time by Johnston and Tiegs in 1922. 

It is significant to note that where, in the compared female, there is only 
one row of disco-lateral bristles on the first abdominal segment, in that of the 
described male and in both sexes of other bred specimens there are two rows. 
These bristles are very well developed, so it excludes the possibility of the second 
row being reduced on the compared specimen. It seems possible that the com- 
pared female belongs to the species known to us as S. alpha, but a further study 
of females, including that of Walker's type specimen, is necessary before any 
finality can be reached concerning the true identity of Walker's species. 

Description. — d". Head. Outer verticals slightly longer than post-oculars; 
frontals twelve on one side, thirteen on the other; facials two, beyond which there 
is a line of about ten hairs; four orals; one row of post-oculars. 

Thorax. Two intra-alars and a third anterior to the transverse suture ; 
presutural and preseutellar acrostichals present. 

Abdomen. On the first segment two I'ows of five or six discal laterals, 
and one strong and one weak submarginal. On the second, one strong and one 
weak lateral submarginal. On the third, one median pair and three lateral sub- 
marginals. On the fourth segment six pairs of submarginals. 

Genitalia. Forceps long. Anterior clasper with a broad flange, both 
claspers of about equal length. The apex of the anterior appendage with a con- 
spicuous, forwardly directed, hook-like process that is characteristic of the species; 
the apical process paired; the lobe is very distinct; filaments present, projecting 
just beyond the apex. 

Legs. The chaetotaxy conforms to the general type, but with the addition 
of a second dorsal row represented by two adjacent bristles at about two-thirds 
the length on the posterior femur; and the second ventral row consists of weak 
bristles. Long hair occurs on all femora and on posterior tibiae. 

V. Eleven frontals on one side, twelve on the other; one long stout facial 
and about five further small ones; one or two stout orals and three or four hair- 
Like ones. A single row of four or five strongly developed disco-laterals. In 
other respects the chaetotaxy conforms to that of the male. 
Hah. — Queensland; New South Wales; Tasmania. 

Observations. — The above description is taken from the allotype male and 
from the female that was compared by Major Austen with the holotype female 
in the British Museum. As pointed out above, these very probably represent two 
species, and it is impossible at present to determine which, if either, belongs to 
the true S. impatiens of Walker. 

The female of S. impatiens that was described by Johnston and Tiegs has 
characters corresponding to the allotype male and in our key to females we have 
ignored the characters of Major Austen's compared female; we leave the true 
identity of the species for further consideration. If tlie characters of the com- 
pared specimen were taken into account, then tlie female would come into the 
beta-omikron group of the key. 

Sarcophaga kpsilon Johnston and Tiegs. (Text-fig. 11.) 

Sarcopharja epsilon, Johnston and Tiegs, Rec. Austr. Mus., xiii., 1922, p. 180, 
fi''. 1. 



BY T. HAKVEY JOHNSTON AND G. H. HARDY. 



Ill 



Description. — 3. Head. Outer vertical bristles slightly longer than the 
post-oculars; twelve frontals; two strong and four short facials; eleven orals, 
beyond which there is a line of bristle-like hairs; three rows of post-oculars. 

Thorax. Two intra-alar bristles ; acrostichals absent (the thorax is badly 
denuded on the holotype but two other specimens examined show that the acros- 
tichals are absent, or, if present, not discernible from the hair, which is rather 
long and plentiful on this species). 

Abdomen. On the first segment one row of five discal lateral bristles 
and three submarginals. On the second, two lateral submarginals. On the third, 
one median pair and three lateral submarginals. On the fourth segment six pairs 
of submarginals. 

Genitalia. Forceps slightly curved. Anterior clasper with a well deve- 
loped flange; the two claspers of about the same length. Anterior appendage 
very small and simple; lobe inferiorly defined; lateral process probably repre- 
sented by a pair of appendages that lie close to the equally long apical process; 
filaments long. 

Legs. The chaetotaxy differs from the general type by the absence of the 
second ventral row on the posterior femur. Long hairs occur scantily on the 
anterior femur and posterior tibiae. 

Hafe.— Queensland. New South Wales: 2 <S <S, Collaroy, Sept., 1921. 

Otservations. — The description is taken from the holotype male in the Aus- 
tralian Museum. 



Sarcophaga haedyi Johnston and Tiegs. (Text-fig. 12.) 

Sarcophaga hardyi, Johnston and Tiegs, Rec. Austr. Mus., xiii., 1922, p. 180, 
PL xxxv., fig. 5. 

Description. — d*. Head. Outer verticals about the length of the post- 
oculars; frontals eleven on one side, twelve on the other; two rows of about 
seven facials; five orals; one row of post-oculars. 

Thorax. Three intra-alar bristles and a fourth placed presuturally ; pre- 
sutural and prescutellar acrostichals strong. 

Abdomen. On the first segment one row of three diseal lateral bristles 
and one submarginal. On the second, one lateral submarginal. On the third, one 
median pair and one lateral submarginal. On the fourth segment, five pairs of 
submarginals alternating with marginals. 

Genitalia. Forceps rather short. Anterior clasper longer than posterior 
and with a short but broad flange near the base. Second joint of the penis com- 
posed mostly of thin, uniformly chitinised plates; anterior appendage broad and 
large, containing an anterior transparent process that is difficult to detect and 
terminating in a row of spines; lobe produced into a rather long triangular pro- 
cess; lateral process curved under a broad flange situated on the apical process; 
apical process, besides containing a pair of these fl.anges, is bifid at the tip; fila- 
ments present. 

Legs, The chaetotaxy conforms to the general type but the second ventral 
row on the posterior femur is weak. Long hair occurs on all femora and on 
posterior tibiae. 

?. The chaetotaxy of the female corresponds remarkably well with that of 
the male; there are ten frontal bristles. 

Hah. — Tasmania. New South Wales: Kosciusko, two males taken by Dv. E. 
W. Ferguson, December, 1921. 






B 



.r> 



112 



AUSTRALIAjST DIPTERA belonging to the genus SAROOPHAGA; 



Observations. — The description is taken from the holotype and the allotype 

in the Australian Museum. 

» 

Sarcophaga misera Walker. (Text-fig. 13.) 

Sarcopliaga misera, Walker, List Dipt. Brit. Mus., iv., 1849, p. 849 ; Johnston 
and Tiegs, Proc. Roy. Soc. Queensland, xxxiii., 1921, p. 67, fig. 22; Ree. Austr. 
Mus., xiii., 1922, p. 178. 

Status. — Walker's type is a female in the British Museum. Major Austen 
determined by comparison with the type, two female specimens which are un- 
doubtedly identical with those determined later and described by Johnston and 
Tiegs under Walker's name. 




^'P{ fiii l.p. a.a./ 




Text-fig. 12. Sarcopliaga hardyi J. & T. 
Text-fig. 13. Sarcophaga misera Walker. 
Text-fig. 14. Sarcopliaga kohla, n.sp. 



Description. — d*. Head. Outer vertical bristles scarcely as long as the 
post-oculars; eleven frontals; four facials; about fourteen orals; one row of 
post-oculars. 



BY T. HARVEY JOHNSTON AND G. H. HARDY. 113 

T h o r a X. Three intra-alar bristles and a fourth placed presuturaUy ; only 
prescutellar acrostichals present. 

Abdomen. On the first segment, two discal lateral bristles on one side 
and two rows of three each on the other; one submarginal lateral. On the 
second, one lateral submarginal. On the third, one median pair and three lateral 
submarginals. On the fourth segment, six pairs of submarginals. 

Genitalia. Forceps short, simple and slightly curved. The claspei-s of 
equal length. Anterior appendage large, broad; lobe pointed at apex; the lateral 
process long and bifid towards the tip; apical process very small and unpaired; 
filaments j)resent. 

Legs. The chaetotaxy differs from the general type by the absence of the 
second ventral row on the posterior femur. Long hair on intermediate and pos- 
terior femur only. 

$. The chaetotaxy cori-esponds rather well with that of the male; there are 
nine frontal bristles. 

Hah. — Queensland; New South Wales; Victoria; South Australia and Lord 
Howe Island. 

Observations. — The above description is taken from the allotype male and 
the larger of the female specimens that were compared with the holotype female 
in the British Museum by Austen. "We have examined about eighty specimens, 
the majority of which have been recently bred by us. 

Sarcophaga kohla_, n.sp. (Text-fig. 14.) 

Sarcophaga misera var. dux, Johnston and Tiegs, Proc. Roy. Soe. Queens- 
land, xxxiii., 1921, p. 70, fig. 23; Rec. Austr. Mus., xiii., 1922, p. 178. 

Synonymy. — In a note to Dr. E. W. Ferguson, Mr. R. R. Parker referred to 
his S. suhtuherosa as being synonymous with S. dux Thomson and consequently 
Johnston and Tiegs published this information. 

S. misera var. dux, however, cannot be accepted as identical with S. suh- 
tuherosa Parker or >S'. dux Thomson, as there are too many genital and other 
differences between them and, on this account, we describe the species under a 
new name. 

Description. — d*. Head. Outer verticals scarcely longer than the post- 
oculars; frontals ten on one side, eleven on the other; about four facials; twelve 
orals; two rows of post-oculars. 

Thorax. Three inti a-alar bristles and a fourth placed presuturaUy ; pre- 
scutellar acrostichals only present. 

Abdomen. On the first segment, two rows of strong discal lateral bristles 
and one submarginal. On the second, one lateral submarginal. On the third, one 
median pair and two lateral submarginals. On the fourth segment, six paii's of 
submarginals. 

Genitalia. Forceps and claspers similar to those on S. misera. The 
anterior appendage divided into two pairs of more or less equally long and 
narrow parts; lobe conspicuously more prominent than in S. misera; and the 
bifid lateral process is more slender than the corresponding part on that species; 
also the apical process is longer and more pointed. 

Legs. The chaetotaxy conforms to tlie general type. Long hair occurs 
abundantly on all femora and on posterior tibiae. 

Hah. — Queensland; New South Wales. 



114 AUSTRALIAN DIPTERA BELONGING TO THE GENUS SARCOPHAGA^ 

Observations.— Described from one of the original specimens referred to by 
Johnston and Tiegs as S. misera var. dux, and now deposited as holotype in the 
Australian Museum, Sydney. 

We have examined the female specimen from Honolulu, together with a male 
from the same locality that is evidently conspeciflc with it; these agree far better 
with S. misera, although showing characters in chaetotaxy which, if constant, 
would easily separate the Australian S. misera from the Honolulu specimens of 
S. dux. This pair from Honolulu has the second row of black post-ocular bristles 
small and considerably thinned out, every alternate one or two bristles being 
obsolete. Also the penis snows a form approaching S. misera in the shorter 
apical process and the structure of the anterior appendage, the latter approximat- 
ing that drawn by Parker to illustrate his S. suhtuberosa. 

We have examined five specimens of S. kohla, one of which is in the Queens- 
land Museum and two are in Dr. Ferguson's collection. 

Sarcophaga securifera Villeneuve. (Text-fig. 15.) 

Sarcophaga securifera, Villeneuve, Mitt. Zool. Mus. Berlin, iv., p. 123, fig.; 
Bottcher, Deut. Ent. Zeitsch., 1913, pp. 15, 370, fig. 41; Aldrich, Sarcophaga and 
allies, 1916, p. 202, fig. 95. 

Description. — c?. Head. Outer verticals strong, at least twice the length 
of the post-oculars; frontals from seven to twelve; facials about seven; orals 
about twelve; one row of post-oculars. 

Thorax. Two intra-alars; acrostichals absent (in male only). 

Abdomen. On the first segment, one row of discal laterals and one sub- 
marginal. On the second, one lateral submarginal. On the third, one median 
pair and two lateral submarginals. On the fourth segment, five or six pairs of 
submarginals. 

Genitalia. Genital segments red. Forceps rather broad. Claspers 
simple, slender, the anterior longer than the posterior. Anterior appendage 
small, simple; lobe rather sinuous and pointed; lateral process long and ter- 
minating in a small flat semicircular disc; filaments rather complex at the apex, 
their true nature not having been detected. 

Legs. The chaetotaxy appears to conform to the general type. Long hair 
occurs on the anterior femur and on the posterior femur and tibia. 

?. Contrary to the usual rule, this species has several characters on the 
female that disagree with those on the male. The preseutellar acrostichals are 
strong and there are two strongly developed rows of discal lateral bristles on the 
first abdominal segment. The genital segments of the female are red like those 
of the male, a character that will readily separate the species from all others 
described in this paper. 

Hab. — Originally described from the Canary Islands, this species has a wide 
range covering Europe, North America and now the new record from Australia. 
Twelve specimens before us were bred by Dr. E. W. Ferguson from a female 
captured in Sydney during April, 1922, the flies emerging during September and 
October of the same year. 

Observations. — The dates supplied by Dr. Ferguson show that the species 
winters in the pupal stage between twenty-two and twenty-six weeks, whereas 
all other species so far bred by us have remained only twelve or thirteen weeks 



BY T. HARVEY JOHNSTON AND G. H. HAKUY. 



U5 



maximum as a pupa, the one exception being an isolated female iLat persisted in 
the pupal condition for a little under twentj^ weeks. 

This long wintering period, the fact that the locality Sydney is a port in 
direct shipping communication with the United States of America, the abundance 
with which the fly occurs in North America in association with markets and 
Buch-like places, as well as the fact that it is the only species yet recorded from 




Text-fi^. 15. Sarcophaga securifera Villeneuve. 
Text-fig. 16. Sarcophaga aurifrons Macquart. 
Text-fig. 17. Sarcophaga omega J. & T. 

Australia with red genital segments, all point to this species as having been 
introduced into Australia. 



Sarcophaga aurifrons Macquart. (Text-fig. 16.) 

Sarcophaga aurifrons, Macquart, Dipt. Exot. suppl. 1, 1846, p. 191. — S. 
aurifera, Brauer and Bergenstamm, Denkschr. Akad. Wiss. Wien, Iviii., 1891 
{nomen nudum). — S. aurifrons, Brauer, Denk. Akad. Wiss. Wien, Math.- Nat. 
CI., evii., 1898, p. 21; Johnston and Tiegs, Proc. Roy. Soc. Queensland, xxxiii., 
1921, p. 71, fig. 4; Also xxxiv., 1922, p. 183; Ree. Austr. Mus., xiii., 1922, p. 
178. — S. sigma, Johnston and Tiegs, Proc. Roy. Soe. Queensland, xxxiii., 1921, p 
84; Ree. Austr. Mus., xiii., 1922, p. 180. 

Synonymy. — The name Sarcophaga aurifrons has been applied to quite a 
number of distinct species by various Australian economic entomologists In 
1921, Johnston and Tiegs determined the identity of a female specimen identified 



116 AUSTRALIAN DIPTERA BELONGING TO THE GENUS SAR(X)PHAGA_, 

as Maequart's species by Coquillet. A critical re-examination of the genitalia of 
S. Sigma has convinced us that this species belongs to ^S^. aurifrons as identified 
by Johnston and Tiegs and therefore this second name is placed as a synonym. 
The genital character referred to under S. sigma as "laterally there is a curious 
tube-like structure of a pale brown colour" is a misinterpretation due to chitini- 
sation giving the appearance of a tube. 

J Description.- — d*. Head. Outer vertical bristles a little longer than the 
post-oeulars ; f rentals, nine and eleven respectively on the two sides of the holo- 
type S. sigma, seven and eight on the specimen described as S. aurifrons; facials 
three or four; orals eleven; three rows of post-oculars. 

Thorax. Three intra-alar bristles and a fourth placed presuturally ; weak 
prescutellar acrostichals only. 

Abdomen. On the first segment, one row of three or four discal lateral 
bristles and one submarginal. On the second, one lateral submarginal. On the 
third, one median pair and three lateral submarginals. On the fourth segment, 
five pairs of submarginals. 

Genitalia. Forceps short. Anterior and posterior claspers about equal 
in length. Anterior process broad and widely expanding; lobe formed into a long 
process; lateral process very long and slender; apex not, or scarcely, forming a 
definite process. . 

Legs. The chaetotaxy conforms to the general type. Long hair oecurs 
scantily on the intermediate, and abundantly on the posterior, femora. 

?. The chaetotaxy of the female corresponds rather well with that of the 
male. There are eight facials on the holotype of S. sigma and nine on the 
specimen described as S. aurifrons, which specimen has the facials and orals re- 
duced in number. 

Hab. — Queensland; New South Wales; South Australia. 

Observations. — The above description was taken from the holotype and allo- 
type of S. sigma, and from male and female specimens described by Johnston 
and Tiegs as S. aurifrons; all these are in the Queensland Museum. The genitalia 
were drawn from a further specimen marked as being that from which the original 
drawing was taken to figure S. aurifrons and which is also in the Queensland 
Museum. , 

Sarcophaga omega Johnston and Tiegs. (Text-fig. 17.) 

Sareophaga frbggatti, Johnston and Tiegs, Proc, Roy. Soe. Queensland, 
xxxiii., 1921, p. 73, fig. 12; Also xxxiv., 1922, p. 183; Rec. Austr. Mus., xiii., 
1922, p. 179; {nee. Taylor). — S. {Parasarcophaga) omega, Johnston and Tiegs, 
Proc. Roy. Soc. Queensland, xxxiii., 1921, p. 86, figs. 25, 26. 

Synonymy. — As pointed out below under Sarcophaga froggatti Taylor, that 
.species conforms to S. theta Johnston and Tiegs. S. froggatti Johnston and 
Tiegs was based upon a specimen that had been compared by Mr. G. Hill with 
the type material in the Tropical Institute, Townsville, but without comparing the 
male genitalia. 

■S. froggatti Johnston and Tiegs has its genitalia identical with those of S. 
omega, the excrescences on the head of which are due to a persistence of the 
ptilinum that is unusually even on each side in this case, but paralleled by similar 
cases amongst other Muscoidean flies captured by us. 

Aldrich determined a specimen of this fly as S. knahi Parker, but Parker's 
drawing of the genitalia of that species differs from the one we give hero in 



BY T. HARVEY JOHNSTON AND G. H. HARDY. 117 

several respects and therefore it seems advisable to await further particulars be- 
fore accepting Aldrieh's determination which, if correct, would necessitate re- 
verting to Parker's name. 

Description.— (S. Head. Outer verticals apparently absent; nine frontals 
on one side, ten on the other; three or four minute facials; about twelve orals; 
one row of post-oculars. 

Thorax. Two intra-alars ; presutural and preseutellar acrostichaLs present. 

Abdomen. On the first segment, two rows of four discal lateral bristles 
and three submarginals. On the second, two lateral submarginals. On the third, 
six (or seven) pairs of submarginals; on the fourth segment, six pairs of sub- 
marginals alternating with marginals. 

Genitalia. Forceps short. Claspers simple, anterior much longer than 
the posterior. Anterior appendix complex; lobe long, directed inwards; lateral 
process long; apical process and filaments absent. 

Legs. The chaetotaxy differs from the general type by having the second 
ventral row on the posterior femur reduced to hairs. Long hair occurs on all 
the femora and on posterior tibiae. 

Hah. — Queensland. 

Observations. — The description is taken from the holotype of S._ omega in 
the Queensland Museum. All the specimens grouped under the name ^S*. frog- 
gatti in the collections examined were captured and not bred, so on this account 
there are none amongst those revised that can indisputably be placed as the 
female of this species. 

Saroophaga ETA Johnston and Tiegs. (Text-fig. 18.) 

SarcopJiaga eta, Johnston and Tiegs, Proc. Roy. Soc. Queensland, xxxiii., 
1921, p. 65, fig. 14; Also xxxiv., 1922, p. 183; Rec. Austr. Mus., xiii., 1922, p. 
378. 

Description. — d". Head. Outer verticals about as long or slightly longer 
than the post-oculars; eight frontals on one side, nine on the other; two stout 
facials and a line of further hair-like ones; about twelve orals; two rows of post- 
oculars. . 

Thorax. Three intra-alars ; only preseutellar acrostichals present. 

Abdomen. On the first segment, two rows of two discal lateral bristles 
and three submarginal. On the second, two lateral submarginals. On the third, 
one median pair and three lateral submarginals. On the fourth segment, four 
pairs of submarginals. 

Genitalia. Forceps rather long and slender, seen from the rear they 
curve outwards. Anterior clasper slightly longer than the posterior. Anterior 
appendage produced downward into a long process; lobe long and pointed; lateral 
process long and rather thin; apical process small, unpaired; filaments long. 

Legs. The chaetotaxy differs from the general type by having the second 
ventral row on the intermediate and posterior femora reduced to hairs. Long 
hairs occur on all femora and on posterior tibiae. 

9. Four or five orals and only one row of post-oculars. Three intra-alars 
on one side, two on the other, but on each side there is a further one placed pre- 
suturally. The second ventral rows on the intermediate and posterior femora are 
represented by a few bristles. 

Hab. — Queensland. 



118 AUSTRALIAN DIPTERA BELONGING TO THE GENUS SARCOPHAGA^ 

Observations. — Our description is taken from the holotype and allotype 
specimens in the Queensland Museum and it will be noted that the female dis- 
agrees with the male in several respects. 

In the original description this species is stated to be described from speci- 
mens bred from fish, but the allotype bears a label containing "fr. bad meat; 
Brisb. 10/20." 

Other females examined by us, including some recently bred, have the second 
row of post-oculars present, and even a third row, indicated on most males, is 
sometimes definitely indicated on the female; nevertheless, the two rows of disco- 
laterals on the first abdominal segment appear to be a constant feature on the 
female, whilst the male invariably has but one. In the key we have taken ac- 
count of our bred specimens in preference to the characters of the allotype, which 
specimen we consider has been erroneously labelled. 

Sarcophaga froggatti Taylor. (Text-fig. 19.) 

Sarcophaga froggatti^ Taylor, Bull. Ent. Res., vii., 1917, p. 265; Parker, 
Canadian Ent., liv., 1922, p. 8, fig. 4. {Nee Johnston and Tiegs, 1921-2, which = 
omega). — S. theta, Johnston and Tiegs, Proc. Roy. See. Queensland, xxxiii., 1922, 
p. 73, fig. 5; Rec. Austr. Mus., xiii., 1922, p. 179. 

Synonymy. — Since the publication of the description of S. froggatti by 
Johnston and Tiegs, Parker gave another species this name, basing his identi- 
fication on a single paratype female supplied by Taylor. As it became necessary 
to ascertain to which species Taylor's name belonged, Mr. G. F. Hill, of the 
Institute of Tropical Medicine, Tov/nsville, kindly submitted the holotype to us 
for study and to have the genitalia extracted. We find that the paratype 
examined by Parker has probably been correctly associated with the male, but 
the details v>'ith respect to the anterior appendage, the accessory plate and the 
three minute spines towards the apex of the forceps are not in accordance with 
the type. We have drawn the genitalia of Taylor's holotype and this will readily 
be recognised as being the same as 8. theta Johnston and Tiegs. 

Description. — d*. Head. Outer verticals a little or scarcely longer than 
the post-oculars. Nine frontals on one side, eight and ten respectively on the 
other side of the holotypes of S. froggatti and S. theta; four very small facials; 
six orals; one row of post-oculars. 

Thorax. Two intra-alar bristles ; only presutural acrostichals present. 

Abdomen. On the first segment, one row of discal lateral bristles^ one 
strong and one weak submarginal. On the second, one lateral submarginal. On 
the third, one median pair and two lateral submarginals. On the fourth seg- 
ment, six pairs of submarginals alternating with slender marginal bristles. 

Genitalia. Seen from the rear the forceps are curved inwards. Claspers 
of about equal length. Anterior appendage with two paired parts, the inner pair 
being the longer; lobe and lateral process small. 

Legs. The chaetotaxy differs from the general type by the absence of the 
second ventral row on the posterior femur. Long hair on the anterior and pos- 
terior femora only. 

?. The chaetotaxy of the female corresponds rather well with that of the 
male; there are eight frontals on the allotype of S. theta. 

Hafe.— Queensland. 



BY T. HARVEY JOHNSTON AND: G. H. HARDY. 



119 



Observations. — The above description was taken from the holotype of S. frog- 
gatti Taylor in the Institute of Tropical Medicine, Townsville, and from the holo- 
type and allotype of S. theta Johnston and Tiegs, in the Queensland Museum. 
The original illustration of S. theta was taken from a paratype specimen. 

Sarcophaga depressa (Robineau Desvoidy). (Text-fig. 20.) 

Myophora depressa, Robineau Desvoidy, Essai Myod., 1830, p. 353. — Sar- 
cophaga depressa, Johnston and Tiegs, Rec. Austr. Mus., xiii., 1922, p. 197, PI. 
XXV., fig. 4; Proe. Roy. Soc. Queensland, xxxiv., 1922, p. 183. — Myophora musca, 




Text-fig. 18. Sarcophaga eta J. & T. 
Text-fig. 19. Sarcophaga froggatii Taylor, 

A drawing of the genitalia taken from the holotype in the Institute of Tropical 
Medicine, Townsville. 

Portions of the genitalia taken from the holotype of .S". iheta J. & T. in the 
Queensland Museum, and to which the letters are attached. 



Robineau Desvoidy, ibidem, 1830, p. 360. — Sarcophaga flavifemorata, Maequart, 
Dipt. Exot. suppl. 4, 1850, p. 233.—S. iota, Johnston and Tiegs, Proc. Roy. Soc. 
Queensland, xxxiii., 1921, p. 79, fig. 11. 

Synonymy. — The above synonymy is that given by Johnston and Tiegs in 
1922. 



120 AUSTRALIAN DIPTEEA BELONGING TO THE GENUS SARCOPHAGA^ 

Description. — c?. Head. Outer verticals shorter than the post-oculars; 
eight f rentals terminating parallel with the base of the antennae; five facials; 
about ten orals; one row of post-oculars. 

Thorax. Three intra-alars and a fourth placed presuturally on one side 
only; presutural and prescutellar acrostiehals present. 

Abdomen. On the first segment, one row of about four discal laterals 
and one submarginal. On the second, one lateral submarginal. On the third, 
one median pair and three lateral submarginals. On the fourth segment, five 
pairs of submarginals alternating with smaller marginals. The whole ventral area 
of the abdomen is exceptionally hairy. 

Genitalia. Forceps rather long, and seen laterally with a very charac- 
teristic heel-like lobe near the apex. A small flange near the apex of the an- 
terior clasper which is a little longer than the posterior. Anterior appendage 
slender, seen laterally somewhat hidden and scarcely protruding beyond the apex 
of the enveloping part of the joint; lobe long, directed inwards and therefore 
mostly concealed; lateral process very broad at the base, but tapering to a 
slender serrated apex; behind the lateral process a pair of minute processes and 
behind these again a minute unpaired apical process. 

Legs. At about the midcJle of the posterior ventral row on the inter- 
mediate femvir, there are two exceptionally long bristles; in other respects the 
chaetotaxy conforms to the general type. Long hair on the intermediate and 
posterior femora. 

?. The chaetotaxy of the female corresponds rather well with that of the 
male; eight f rentals on one side and seven on the other, extending beyond the 
base of the antennae in the usual manner; onlj'^ one exceptionally long bristle in 
the second ventral row of the intermediate femur. 

iTaft.— Queensland; New South Wales; Victoria; Tasmania; South Australia; 
Western Australia. 

Observations. — The above description is taken from the holotype and allo- 
type of S. iota Johnston and Tiegs. 

Sarcophaga omikron Johnston and Tiegs. (Text-fig. 21.) 

Sareophaga omikron^ Johnston and Tiegs, Proc. Roy. See. Queensland, xxxiii., 
1921, p. 82, fig. 16 ; Rec. Austr. Mus., xiii., 1922, p. 180. 

Description. — d*. Head. Outer verticals scarcely longer than the post- 
oculars; ten frontals on one side, eleven on the other; two rows of about four 
facials each; about six orals; one row of post-oculars. 

Thorax. Two intra-alars ; prescutellar and presutural acrostiehals pre- 
sent, the latter, however, owing to a fracture, are not to be detected on the holo- 
type. 

Abdomen. On the first segment, one row of four discal lateral bristles 
and one submarginal. On the second, one lateral submarginal. On the third, 
one median pair and two lateral submarginals. On the fourth segment, five 
submarginals alternating with marginals. 

Genitalia. Forceps rather simple, curved and diverging towards the 
apex. Cla-spcrs long and strongly curved forward, the anterior one shorter than 
the posterior. Anterior appcnrlage small; the lobe not at all conspicuous; a pair 
of knobs is situated at what may be the true apex, but the lateral processes are 



BY T. HARVEY JOHNSTON AND G. H, HARDY. 



121 



complex and joined at the base by a hood that forms a further process, un- 
paired; filaments presents. 

Legs. The chaetotaxy differs from the general type by the absence of the 
second ventral row on the posterior femur. Long hair occurs on the posterior 
tibiae. 

?. The chaetotaxy of the female corresponds rather well with that of the 
male. Nine frontals; five submarginals on the third abdominal segment. 

Hab. — Queensland; South Australia; Western Australia. 




Text-fig. 20. Sarcophaga depressa Dasvoidy. 
Text-fig. 21. Sarcophaga omikron J. & T. 

Observations. — The description is taken from the holotype and allotype in the 
Queensland Museum. 



Sarcophaga peregrina (Robineau Desvoidy). (Text-fig. 22.) 

Myophora peregrina, Robineau Desvoidy, Essai Myod., 1830, p. 356. — Sar- 
cophaga peregrina, Johnston and Tiegs, Rec. Austr. Mus., xiii., 1922, p. 177; 
Proe. Roy. Soc. Queensland, xxxiv., 1922, p. 182. — Myophora subrotunda, Robi- 
neau Desvoidy, ibidem, p. 357. — M. rapida, Robineau Desvoidy, ibidem, p. 360. — 
Sarcophaga irrequieta, Walker, List Dipt. Brit. Mus., iv,, 1849, p. 830; Johnston 



122 AUSTRALIAN DIPTERA BELONGING TO THE GENUS SARCOPHAGA, 

and Tiegs, Proc. Roy. Soc. Queensland, xxxiii., 1921, p. 63, figs. 1-3. — S. ochri- 
paipis, Thomson, Eugenies Resa, Dipt., 1868, p. 537. 

Synonymy. — The above synonymy relating to species described from Aus- 
tralia, is selected from a longer list of synonymy for this species that was pub- 
lished by Johnston and Tiegs, in 1922. 

Description. — (S. Head. Outer verticals telightly longer than the post- 
oculars; eleven frontals on one side, ten on the other; facials, two rows of about 
five each; about ten orals; three rows of post-oeulars. 

Thorax. Two intra-alars ; prescutellax acrostichals only. 

Abdomen. On the first segment, two rows of about four discal lateral 
bristles and one. submarginal. On the second, one lateral submarginal. On the 
third, one median pair and three lateral submarginals. On the fourth segment, 
about five pairs of submarginals. 

Genitalia. Forceps seen laterally constricted at about two-thirds their 
length, apex broad and terminating in a very short point. Claspers slender, 
the anterior a little longer than the posterior. Anterior appendage with a large 
area of minutely spined structure that is represented in the drawing by a serrated 
edge; lobe large and broad, with two conspicuous projections; lateral process very 
broad and containing a small slender projection at the lower corner of the other- 
wise truncate apex; apical process unpaired and short; arising from just posterior 
to the base of each lobe is an interior process which is readily detected by the 
more or less cylindrical shape when seen ventrally; it is not readily perceptible 
from the lateral aspect, being mostly hidden by other parts of the penis; the 
nature of this process is not yet understood, and it has only been found in this 
species and in S. kappa J. & T. ; in the latter it takes quite a different form. 

Legs. The ehaetotaxy differs from the general type by the absence of the 
second ventral row on the intermediate and posterior femora, and by the absence 
of the posterior subapieal bristle on the posterior femur. Long hair occurs on 
the anterior and posterior femora. 

2. Eleven frontals on one side, nine on the other; three intra-alars and a 
fourth placed presuturally ; two lateral submarginals on the first abdominal seg- 
ment and those on the second obsolete; the ehaetotaxy of the legs conforms to 
the general type; in other respects the female corresponds with the male. 

Hab. — Queensland ; New South "Wales ; South Australia. 

Observations. — The description is based on a male and a female in the 
Queensland Museum, which were selected from the series used in 1921, for the 
account of S. irrequieta Walker by Johnston and Tiegs, who utilised a general 
set of characters that occurred in the majority of the specimens and, consequently, 
their description does not fit any particular pair of specimens in the series. The 
characters of the ehaetotaxy are very variable within this species. 

Sarcophaga BANCROFT! Johnston and Tiegs. (Text-fig. 23.) 

Sarcophaga bancrofti, Johnston and Tiegs, Proc. Roy. Soc. Queensland, 
xxxiii., 1921, p. 85, fig. 8. 

Description. — 6. Head. Outer verticals obsolete on one side and repre- 
sented by a very short bristle on the other; nine frontals; one long and three other 
facials; nine or ten orals; two rows of post-oculars. 

Thorax. Three intra-alars, and a fourth placed presuturally; prescutcllar 
acrostichals present, the pref=utural acrostichals, if present, are not to be dis- 
tinguished from the bristly vestiture around them. 



BY T. HARVEY JOHNSTON" AND G. H. HARDY. 



123 



and 

live 



iut'- 



Abdomen. On the fii'st segment, one row of diseal lateral bristles 
one submarginal. On the second, two lateral submarginals. On the third, 
pairs of submarginals. On the fourth segment, six pairs of submarginals. 

Genitalia. Forceps very short and rather broad. Claspers not 
ficiently perceptible to be drawn, but they are simple and of about equal length. 
Second joint of the penis exceptionally small; anterior appendage folded so that 
the apex is directed towards the base; lobe present and directed inwards; apical 
process small and paired. 




Text-fig. 22. Sarcophaga peregrina Desvoidy. 
The second joint of the penis (E) has the interior process omitted in the lateral 
view; this process is mostly hidden and arises where indicated by the dotted line 
from ' ' i.p. ' ' 

Text-fig. 23. Sarcophaga bancrofti J. & T. 
Text-fig. 24. Sarcophaga fergusoni J. & T. - 

The second lateral view (I) of the penis is taken from a specimen that is considered to 
be S. fergusoni, and it shews the true shape of the joint which is evidently damaged on the 
holotype. x indicates an outgrowth occurring on one side only of the holotype and 
apparently accidental. 



Legs. The chaetotaxy conforms to the general type. At most long hair is 
scantily represented on the posterior femur. 

Kab. — Queensland. 

Ohservations.- — The description is taken from the unique holotype in the 
Queensland Museum. 



124 AUSTRALIAN DIPTERA BELONGING TO THE GENUS SARCOPHAGA, 

Sarcx)phaga fergusoni Johnston and Tiegs. (Text-fig. 24.) 

Sarcophaga fergusoni, Johnston and Tiegs, Proc. Roy. Soc. Queensland, 
xxxiv., 1922, p. 186, fig. 2. 

Description. — 6'. Head. Outer vertical bristles about the length of the 
post-oculars; ten f rentals; about seven facials; ten orals; two rows of post- 
oculars, the second of which consists of very weak bristles. 

Thorax. Three intra-alars and a fourth placed presuturally ; preseuteJlar 
acrostiehals onh^ 

Abdomen. On the first segment, two rows of discal lateral bristles and 
two submargiaals, but all these are difficult to detect on the holotype. On the 
second, three lateral submarginals. On the third and fourth segments, six sub- 
marginals. 

Genitalia. Like that of ^S*. hancrofti ; the penis is very small and we 
have found it impossible to detect the true nature of the parts on the holotype 
and therefore the following characters are augmented from a second specimen of 
this species. 

Forceps small and more slender than those of 8. hancrofti. Claspers simple 
and of uniform length. Anterior appendage with the apex curved round so as 
to point towards the base; lobe as in S. hancrofti, difficult to detect and directed 
inwards; apical process larger than in S. hancrofti and unpaired. 

Legs. The chaetotaxy conforms to the general type. Long hairs on all 
femora and on posterior tibiae. 

Hab. — New South Wales. The second specimen referred to is from TJralla 
and is dated 25/11/1914. 

Observations. — Described from the holotype in the Australian Museum, ex- 
cept the penis which is evidently mutilated on the holotype. In the drawings we 
have given an outline figure of the penis taken from the holotype, as well as the 
details taken from a second specimen. 

Sarcophaga kappa Johnston and Tiegs. (Text-fig. 25.) 

Sarcophaga kappa, Johnston and Tiegs, Proc. Roy. Soc. Queensland, xxxiii., 
1921, p. 81, fig. 7; Rec. Austr. Mus., xiii., 1922, p. 180.— S. illingworthi, Parker, 
Canadian Ent., liv., 1922, p. 7, figs. 3 & 5. 

Synonymy. — -Mr. R. K. Parker informs us that his species, S. illingworthi, 
is the same as S. kappa. 

Description. — d*. Head. Outer verticals about the length of the post- 
oculars; nine frontals; four or five small facials; six orals; one row of post- 
oculars. 

T h o r a x. Three intra-alars on one side, tw^o on the other ; presutural and 
presr-utellar acrostiehals present. 

A b d in e n. On the first segment, one discal lateral and one submarginal. 
On the ser-oiid, one lateral submarginal. On the third, one median pair and two 
lateral submarginals. On the fourth segment, five or six pairs of submarginals. 

Genitalia. Forceps rather long and slightly curved. Anterior and pos- 
terior claspers short and of about equal length. Anterior append- 
age small-, containing a pair of widely separated and well defined short processes; 
lobe moderately long; apical process consisting of a pair of broad flanges; two 
closely adjacent interior processes extend a little beyond the apical processes; 
they are referred to under the description of S. peregrina. 



BY T. HARVEY JOHNSTON AND G. H. HARDY. 125 

Legs. The chaetotaxy differs from the general type by having onJy one 
ventral row on the posterior femur. Long hair on all femora and on posterior 
tibiae. 

5:^. Nine f rentals on one side, ten on the other; two rows of two discal 
laterals on the first abdominal segment; the chaetotaxy of the femora conforms 
to the general type; in other respects the female conforms to the male. 

Hub. — Queensland; New South Wales. 

Observations. — The description is taken from the holotype and allotype in 
the Queensland Museum. The one isolated discal bristle on the first abdominal 
segment is probably due to a reduction of bristles occurring on both sides, where 
normally the male as well as the female should have two rows of these bristles; 
similar eases have been found on individual specimens of other species that nor- 
mally have two rows. 

Saroophaga gamma Johnston and Tieg-s. (Text-fig. 26.) 

Sarcophaga gamma, Johnston and Tiegs, Proc. Roy. Soc. Queensland, xxxiu.., 

1921, p. 60, fig. 15; Also xxxiv., 1922, p. 182; Ree. Austr. Mus., xiii., 1922, p. 
180. — S. brunneopalpis, Johnston and Tiegs, Proc. Roy. Soc. Queensland, xxxiv., 

1922, p. 184, fig. 1. 

Synonymy. — We have recently secured a series of about forty specimens of 
this species and have concluded that despite the difference in chaetotaxy, and the 
supposed difference in genitalia between the two forms, S. gamma and S. brun- 
neopalpis, the two really represent one species. The difference in the genitalia 
between the two forms will be found near the apex of the second joint of the 
penis, where the process-like lobe and the filaments are clearly discernible in one, 
but hidden and apparently absent in the other. We have observed several speci- 
mens that exposed these parts when first the genitalia were extracted, but the 
penis became distorted on drying and thus came to conform to the "gamma" 
type. Moreover, there are both forms of genitalia within the series, with and 
without the corresponding presutural acrostichal bristles. 

Mr. R. R. Parker informs us that he considers S. gamma Johnston and 
Tiegs, is identical with S. ochidea Bottcher, but Bottcher's drawing of the geni- 
talia differs from ours in several respects, the most noticeable of which will be 
found in the relative size and shape of the claspers. Bottcher's localities include 
New Guinea, nevertheless we consider it advisable to await further information 
before accepting this possible synonymy. 

Description. — d*. Head. The outer verticals about twice the length of the 
post-oculars; eight frontals, ten on one side in S. brunneopalpis ; one or two 
stout facials; about twelve orals; one row of post-oculars. 

Thorax. Two intra-alar bristles in iS*. gamma, three in S. brunneopalpis; 
presutural and prescutellar acrostiehals, the former, however, missing in the holo- 
type of S. gamma and in many other specimens of our long series. 

Abdomen. On the first segment, two rows of eight discal lateral bristles, 
five in brunneopalpis, and three submarginals. On the second, three lateral sub- 
marginals, two in brunneopalpis. On the third, one median and three lateral sub- 
marginals on S. brunneopalpis, but supplemented to form six pairs on S. gamma. 
On the fourth segment, five or six pairs of submarginals that alternate with 
slenderer marginals. 

Genitalia. Forceps long. Anterior clasper much longer than the pos- 
terior. Anterior appendage widely expanding; lobe forming a smaU process; 



126 



AUSTKALIAX DIPTERA BELONGING TO THE GENUS SARCX)PHAGA, 



apical process paired; filaments present; the apex of the penis has contracted on 
the holotype of S. gamma, so that the lobe and filaments are not apparent on this 
specimen. 

Legs. The ehaetotaxy differs from the general type by having the second 
ventral row on the posterior femur reduced to hairs. Long hair abundant on the 
intermediate and posterior femora and on the posterior tibiae. 

Hah. — Queensland; New South Wales. 

Observations. — The description is taken from the holotypes of S. gamma 
and S. brunneopalpis in the Queensland Museum and we have examined about 
forty further specimens mostly taken around dead birds on Mt. Coot-tha, Bris- 




Toxt-fig. 2.5. Sarcophaga kappa J. & T. 

Text-fig. 26. Sarcophas^a gamma J. & T. 
The apex of the penis from the holotype of 5". brunneopalpis J. and T. (I) shows 
the true form of the lobe and filaments which characters are concealed in the 
holotype of S. gamma. 

bane, during September, 1922. Although males were very abundant in this 
locality, we were not able to associate with the species any female specimens 
also taken during this month. 



Sarcophaga littoralis Johnston and Tiegs. (Text-fig. 27.) 

Sarcophaga littoralis, Johnston and Tiegs, Rec. Austr. Mus., xiii., 1922, p. 
183, PI. XXV., fig. 2; Proc. Roy. Soc. Queensland, xxxiv., 1922, p. 184. 

Description. — S. Head. Outer verticals not traceable, frontals weak^ 



BY T. HARVEY JOHNSTON AND G. H. HARDY. 127 

twelve or thirteen, scarcely extending beyond the base of the antennae; about 
four stout facials; four orals, also stout; one row of post-oculars. 

Thorax. Badly denuded ; probably there are three intra-alars and both 
presutural and prescutellar acrostiehals present. 

Abdomen. On the first segment, one row of about three discal laterals 
and one submarginal. On the second, one lateral submarginal. On the third, 
one median pair and two lateral submarginals. On the fourth segment, seven 
pairs of submarginals alternating with marginals. 

Genitalia. Forceps broad, terminating in a pointed apex. Claspers 
simple and small, the anterior longer than the posterior. Anterior appendage 
small; lobe plainly discernible; the apex consists of a pair of broad flanges, the 
lower edge of each containing a small fold; the flanges are arranged hood-like, 
so that the interior of the penis cannot be seen from the ventral aspect. 

Legs. The chaetotaxy differs from the general type by the absence of the 
subapical bristles on the posterior femur. Long hair occurs scantily on all femora. 

?. Frontals eleven or twelve. As in the male, the thorax is damaged, but 
there are probably three intra-alars, and a fourth placed presuturally ; both pre- 
sutural and prescutellar bristles may be present. The intermediate legs are miss- 
ing; the posterior femur contains one anterior row of bristles, one subapical 
dorsal bristle and at least one ventral row. 

Hab. — Queensland. 

Observations. — The above description is taken from the holotype and allotype 
in the Australian Museum; both specimens are in a very dilapidated condition. 

Saecophaga synia, n.sp. (Text-fig. 28.) 

Description. — c?. Head. The outer vertical bristles scarcely longer than 
the post-oculars; ten frontals on one side, eleven on the other; two strong and 
four or five weak facials; ten orals; three rows of post-oculars, but the two 
posterior ones are formed of rather slender bristles. Third joint of antenrae 
slightly longer than twice the length of the second. Frontal stripe as wide as 
parafrontals. 

Thorax. Three intra-alars and a fourth placed presuturally ; presutural 
and prescutellar acrostiehals present. 

Abdomen. On the first segment, a row of four discal lateral bristles and 
one submarginal. On the second, one lateral submarginal. On the third, one 
median pair and three lateral submarginals. On the fourth segment, five pairs 
of submarginals. 

Genitalia. Forceps rather simple but terminating in a double point. 
Anterior claspers with a small flange near the apex, the posterior smaller and 
simple. The second joint of the penis recalls that of S. carnaria Meigen; the 
anterior appendage consists of a pair of bowed processes that almost meet at 
the apex; the lobe is very well differentiated; the apex curves forward like that 
of S. carnaria and is unpaired; filaments present. 

Legs. The chaetotaxy differs from the general type by the number of 
bristles in the ventral rows on the intermediate femur being reduced, and by the 
anterior and the second ventral row on the posterior femur being obsolete. 

Colour. — Silvery grey, with slight indications of yellowish in places. 

Length. — 12 mm. 



128 



AUSTRALIAN DIPTERA BELONGING TO THE GENUS SAROOPHAGA, 



Hob. — Queensland: Mt. Coot-tha, Brisbane, September, 1922; one male, 
unique. 

Observations. — This species bears a remarkable resemblance to S. camaria 
Meigen in regard to the characters of the male genitalia; in both species the 
forceps and elaspers are of about the same length and of the same general form; 
the penis also agrees in having characters in common. Possibly this species is 
the original one upon which the record of S. camaria in Australia was based. 

The specimens of S. carnaria Meigen examined by us consists of two males 
and a female from Europe; a male determined by C. J. Wainwright in 1915, and 
a pair determined by Bottcher and kindly supplied by Dr. M. Bezzi. 




Text-fig. 27. Sarcophaga lUtoralis J. & T. 
Text-fig. 28. Sarcophaga synia, n.sp. 



Unrecognised Species. 
Sarcophaga ptaedatrix Walker. (List Dipt. Brit. Mus., iv., 1849, p. 826). 

Status. — This species was described by Walker as having the third antennal 
joint three times the length of the second; and the length of the body five lines. 
The remainder of the account would fit almost any species of the genus. It was 
evidently described from a moderately large specimen, not particularly "golden," 
in which the third antennal joint is longer than usual and such characters would 
conform to the female of S. kappa Johnston and Tiegs. 

Locality. — Port Essington, Northern Territory. 

Sarcophaga pallichrus Thomson. (Eugenics Resa, 1869, p. 539). 

Status. — This species has been placed provisionally by Van de Wulp under 
Sarcophagula. Its small size suggests Helicohia australis, but the original das- 
Bcription seems to indicate that it is not a Sarcophagid fly. 



BY T, HARVEY JOHNSTON AND G. H. HARDY. 129 

List of works referred to. 

Aldrich^ 1916. — Saxcophaga and allies in North America. 

BoTTCHER^ 1912-13. — Die mannliehen Begattungswerkzeuge bei dem Greuus 

Sareophaga Meigen, und ihre Bedeutung fur die Abgi-enzung 

der Arten. Deut. Ent. Zeit., Berlin. 
, 1913. — Sauters Formosa-Ausbeute. Einige neue Sarcophaga-Arten. 

Ann. Mus. Nat. Hungariei, si., pp. 374-381. 
Brauer^ 1898. — Beitrage zur Kenntniss der Muscaria scMzometopa. Sitz. k. Akad. 

Wiss., math.-nat. CI. AhtJi., cvii. 
Graham-Smith, 1914. — Flies and Disease; non-bloodsucking flies. Second edi- 
tion, p. 35 (recording Sareophaga carnaria L. from Australia). 
GuERiN^ 1830. — Voyage de la Coquille; zoologie ii. 
Johnston and Haudy, 1923. — Some Sarcophagid flies from Lord Howe Island. 

Bee. Aust. Mus., xiv. (1), 62-71. 
Johnston and Tiegs, 1921. — New and little known Sarcophagid flies from South 

East Queensland, Proc. Boy. Soc. Queensland, xxxiii., pp. 46- 

90. 
, 1922. — Sarcophagid flies in the Australian Museum Collection. 

Bee. Aust. Mus., xiii., pp. 175-188. 
, 1922. — New and known Australian Sarcophagid flies. Proc. Boy. 

Soc. Queensland, xxxiv., pp. 181-190. 
Macquart, 1846, 1850. — Dipteres exotiques nouveau ou peu connus; supplements 

1 et 4. Reprinted from Mem. Soc. roy. Sci., Agric. et Arts, 

Lille, 1838-54. 
Parker, 1922. — Australian Sarcophagidae ; new species and data concerning 

others. Canad. Entom., liv., pp. 4-9. 
RoBiNEAU Desvoidy, 1830. — Essai sur les Myodaires. Mem. Acad. Sav. Strangers, 

ii., Paris. 
Taylor, 1917. — Sareophaga froggatti, a new sheep-maggot fly. Bull. Ent. Bes. 

London, vii., p. 265. 
Thomson, 1868. — Sweden. Kongl. Sv. Freggatten Eugemies Besa, etc., ii., zooL, 

part 12, Dipt. 
TowNSEND, 1917.^ — Genera of the Dipterous tribe Sarcophagini. Proc. Biol. Soc, 

Washington, xxx., pp. 189-197. 
ViLLENEUVE, 1908. — In Becker. Dipteren der Kanarischen Inseln. Mitt. Zool. 

Mus., Berlin, iv., pp. 121-126. 
Walker, 1849. — List of the Dipterous insects in the British Museum, iv. 



130 



THE LOBANTHACEAE OF AUSTRALIA. Part iv. 

By W. F. Blakely, Botanical Assistant, National Herbarium, Sydney. 

(Plates iii.-xiv.) 
[Read 18th April, 1923.] 

9. LoRANTHUS QUEENSLANDicus^ n.sp. (Plate iii.) 

Frutex ramis giabris robustis teretibus vel apud nodos minime compressis. 
Folia opposita late-spatbulata ad elliptica plana glabra superne nitentia sub- 
eoriacea 3-5 costata in semi-tereti petiolo attenuata 4-7 cm. x 3-4 cm. Inflores- 
centia cj'mosa eymis axillaribus eorymbosis communi pedunculo crasso 3-5 cm. 
longo 3-6 radiato utroque radio cymum 3 florum flavorum quorum centralis 
sessilis ferente. Gremmae cylindracea 30-35 mm. faciliter in 6 libera linearia seg- 
menta separatae. Antberae adnatae lineares 4 mm. Stylus gracilis subangularis 
37 mm. Stigma miniorum. Fructus urceolatus vel ellipticus glaber 8-12 mm. 

Branches glabrous, robust, terete, or slightly compressed at the nodes. 
Leaves opposite, broad spathulate, elliptical, flat, smooth and shining above, dull 
beneath, 4-7 cm. long, 3-4 cm. broad, thin, coriaceous, 3 — 5-nerved, the numerous 
fine lateral nerves indistinct, narrowed into a semiterete petiole 6-10 mm. long'. 
Flowers yellowish-brown, in axillary sub-corymbose cymes; the common peduncle 
rather thick, 3-5 cm. long, bearing 3-6 rays, each with a partial cyme of 3 flowers 
aiTanged in triads, the central flower sessile, the lateral flowers on stout, bracteate 
pedicels. Bracts of the sessile flowers rather large, concave, lanceolate, gibbose; 
bracts of the lateral flowers smaller, broad spathulate to elliptical, or nearly orbi- 
cular, tiTincate, minutely rusty tomentose inside and on the margins, much en- 
larged under the fruit. Calyx ))road eupular, the limb somewhat prominent, 
slightly contracted, truncate or minutely ciliate-denticulate. Buds cylindrical, 30- 
35 mm. long, readily separating into 6, free, narrow linear segments when mature ; 
petals acute, bearded inside at the apex with a tuft of minute brown deciduous 
hairs. 

Filaments very narrow, 9 mm. long; anthers adnate, linear, *4 mm. long. 
Style 37 mm. long, slender, sub-ang-ular ; stigTQa very small. Disc flat. Fruit 
urceolate to elliptical, smooth, 8-12 mm. long, but not seen ripe. Epicarp thick. 

Range. — This species is at present confined to the north-east coast of Queens- 
land. 

Barron River, native name "Darandool" (E. Cowley, No. 5c. in Queensland 
Herbarium. Recorded as L. diclyophlehus in Bailey's "Comprehensive Catalogue 
of Queensland Plants," p. 460); Herberton (S. Dixon. The type); Cairns (F. 
M. Bailey, in Queensland Herbarium). 



BY W. ¥. BLAKELY. 131 

Affinities. — L. Queenslandicus is very closely allied to L. harbeUutm Blakely, 
a New Guinea species, from which it differs in the less coarse and more obscurely 
veined leaves; larger peduncles and pedicels; more variable In-acts; longer and 
thicker buds; and in the filaments being twice the length of the anthers; also in 
the depressed disc, and rusty vestiture of the calyces and bracts. 

It shows affinity with L. congener, especially the short broad elliptical forms 
of the latter, but it is much larger than L. congener in all its parts. 

Host not stated. 

10. LORAisTTHUS CYCJfEUS-SiN-us^ n.sp. (Plate iv., fig. A.) 

Frutex giaber ramis brevibus robustis divaricatis, lenticulatis, quorum nodi 
laeves paulis per prominentes. Folia opposita, crassa, coriacea., elliptica ad orbi- 
eularia petiolata obscure 3-5 costata, 3-5 cm. longa, 20-25 mm. lata. Cymi 
axillares, deflexi, bis vel tes ramati, utroque ramo portante tres flores, quorum 
centralis sessilis. Gemmae gracilis clavatae unctae 15-20 mm. longae. Bracteae 
obtusae aequantes dimidium calycem. Calyx urceolatus, laminis ciliatis mem- 
branaceis; segmentis 5, liberis. Antherae oblongo-ellipticae filamentis tenuibus 
adnatae. Stylus angularis; stigma magnum, capitatum. Fructus non visi. 

Glabrous, branches divaricate, firm, lenticulate, with somev.^hat prominent 
smooth nodes. Leaves opposite, thick, coriaceous, broadly elliptical to almost 
orbicular, petiolate, obscurely 3 — 5-nerved, 3-5 cm. long, 20-25 mm. broad. In- 
florescence cymose. Cymes deflexed, 2 — 3-branched or more, the common peduncle 
slender, 5-8 mm. long. Flowers in triads, the central one sessile. Buds slender, 
clavate, greasy, 15 mm. long. Bracts obtuse, about half the length of the calyx. 
Calyx urceolate, 2 mm. long, the membranous limb ciUate, entire or minutely 
toothed. Segments 5, free to the base. Anthers oblong elliptical, adnate to the 
rather thin filaments. Style angular in the lower half, thicker than the filaments; 
stigma large, capitate, minutely verrucose. Fruit not seen. 

Range. — Cygnet Bay, West Kimberley (W. Y. Fitzgerald, labelled L. pen- 
dulus Sieber). 

Affinities. — This species is more closely allied to L. miraculosus Miq., than 
to any other; it is separated by the more coriaceous tri- or quinque-nerved leaves, 
thicker buds, differently shaped calyx and larger stigma. 

In foliage it resembles the elliptical forms of L. congener Sieber, but varies 
in other characters. 

Host not stated. 

11. LoRANTHUs Mackayensis, n.sp. (Plate iv., fig. B.) 

Folia opposita juniora in superfine ventrali ferruginea baud tamen tomen- 
tosa, elliptica ad obovata, flaccida, opaca, in petiola obvio eonstricta, obscure 
tripKnervata, 2-4 cm. longa, 15-20 mm. lata. Flores in cymis axillaribus, pedun- 
culo communi deflexo perente umbellum trirem radiorum, quorum utroque tribus 
floribus, central! sessili. Gemmae acuta clavatae tenues 10-17 mm. longae. Brac- 
teae acutae, minute ciliatae, calycem non supereminentes. Calyx cylindricus, 
vestice constricto deem in gemma, repandus segmentis cadentibus, laminis bre- 
vissimis minute ciliatis 2.5 mm. longis; seg-mentis 5, omnino liberis; antherae 
oblongae adnatae, 2 mm. longae. Stylus tenuis angularis; stig-ma parvum. Fruc- 
tus non visi. 

Glabrous shrubs with short divaricate branches. Leaves opposite, the young 
leaves ferruginous underneath but not tomentose, elliptical to obovate, thin opaque, 
rounded or contracted at the base into a distinct petiole 3-1 mm. long, obscurely 



l32 THE LORANTHACEAE OE AUSTRAliIA_, iv., 

triplinervecl, 2-4 cm. long, 15-20 mm. broad. Flowers in axillary cymes, the 
common peduncle deflexed, about 10 mm. long, bearing an umbel of 2 or 3 rays, 
each with 3 flowers, the central flower sessile. Buds very slender, elavate, acute, 
10-17 mm. long. Bracts acute, minutely ciliate, not exceeding the calyx. Calyx 
cylindrical, contracted at the top in bud, expanded when the petals fall, the 
limb very short, minutely ciliate, about 2^ mm. long. Petals 5, free for their en- 
tire length. Anthers oblong, about 2 mm. long. Style slender, angular; stigma 
very small. Fruit not seen. 

The material is imperfect and does not allow a full description. 

Range. — Mackay, North Queensland (H. Try on). 

Affinities. — Allied to L. miraeulosus in general appearance but diverging in 
the more elliptical, non-glaucous leaves, slender buds and cylindrical calyces. 

L. Cycneus-Sinus resembles this species in the shape of the leaves, which are, 
however, thicker, and usually quinque-nerved ; while the buds are thicker, more 
elavate, and the stigma considerably larger. The calyx, also, is more urceolate 
with a correspondingly larger limb. 

L. Mackayensis also bears some resemblance to L. ohliqua, notably in the 
common peduncle, bracts, and to some extent in the buds. 

Host not stated. 

12. LOEANTHUS MIRACULOSUS Miq. (Plate v.) 

Lehm., Plantae Preiss., 1844, p. 281. 

The following in a translation of the original description: — Branches 
glabrous, terete, internodes usually compressed. Leaves opposite, shortly petio- 
late, fleshy or leathery, elliptical or subspathulate-lanceolate, contracted at the 
base, the apex obtuse, veinless or obscurely one-nerved. Cymes shortly peduncu- 
late, two or three branched, each branch bearing three flowers, the central one 
sessile. Bracts rounded-ovute or navicular, the apex ciliate. Calyx truncate. 

Parasitic on Santalum sp., near Wicklow, 8th March, 1841 (Herb) Preiss., 
No. 1607, in flower, near York, 12th April, 1840; No. 1610, without flowers. 

A large species similar to the host plant; showing remarkable manner of 
parasitism and natural physiology worthy of note. Not only are the leaves the 
same shape, structure and venation, but the young branches are flattened, the old 
ones terete, and of a similar nature to the shrubs of Santalum on which it grows, 
but a little yellower, glabrous, dichotomously branched; branches smooth; branch- 
lets usually dilated or compressed at the top. Leaves opposite or slightly alter- 
nate, leaving a scar as they fall off, thick, earnose, coriaceous, elliptical or spathu- 
late-lanceolate, obtuse or rounded at the top, contracted into a cuneate base 2^-44 
cm. long, 1-1 J cm. broad, nerveless or obscurely 1-nerved; petiole channelled 
above, 3-4 mm. long. Cymes corymbose, axillary and terminal, shorter than the 
leaves, two- or three-branched, frail; peduncles triflorous; flowers pedicellate, or 
the central one sessile. Bracts fleshy, rounded-ovate or boat-shaped, concave, the 
apex minutely ciliate. Calyx obconic, repand-truncate, 2 mm. long, or smaller. 
Petals 5, brown when dry, linear spathulate, the apex elliptical, concave, 8-9 mm. 
long. Stamens small, inserted towards the middle of the petal; anthers linear- 
elliptical. Style filiform, slightly exceeding the stamens, angular or sulcate, 
flexuose in the upper poi'tion ; stigma verrucose, sub-capitate. 

Supplementary notes to the description. 
Glabrous and somewhat glaucous plants with erect or pendulous branches 1-2 



BY W. F. BLAKELY. 133 

feet long, often forming dense masses up to 3 feet in diameter. Union ball-like, 
often larger than in most species, exceeding 6 inches in diameter. Bark on the old 
stem dark and rough, similar to that of Exocarpus cupressiformis. 

Leaves opposite, thick, very variable, elliptical, spathulate, cuneate and 
lanceolate sometimes on the same branch, 2-6 cm. long, 1-2 cm. broad, in the 
typical form, venation imperfectly triplinerved or somewhat penninerved, ap- 
proaching that of L. vitellinus, occasionally three nerves more conspicuous than 
the others, but frequently the central nerve alone conspicuous, petioles 5-10 mm. 
long, terete or compressed. Cymes axillary or terminal, 2 — 4-branehed, usually 
shorter than the leaves, the common peduncle 10-15 mm. long. Buds robust, 
usually crimson, or Old Carmine, clavate, or angular at the top and slightly 
swollen at the base; quite glabrous; only the small ovate bract minutely rufus 
ciliate. Calyx obconical to cylindrical, the conspicuous broad membranous limb 
irregularly denticulate-ciliate. Flowers 4 — 5-merous, 15-20 mm. long. Petals 
free, dark red at the base shading into orange-red upwards, the apex acute. Style 
slender, geniculate, on a level with the base of the anthers, of a dark-red colour 
throughout, and pentagonal or deeply furrowed for its size. Stigma rather large, 
somewhat capitate. Anthers oblong, 3-4 mm. long. Fruit oblong, 8 mm. long, 
about 4 mm. in diameter, contracted at the top, usually a light yellow. Seeds 
elliptical, 5 mm. long, with 5 scarcely perceptible furrows. Endosperm green, 
hypocotyl terete, minutely verrueose, green. Embryonic cotyledons oblong- 
obtuse, scarcely 3 mm. long. On germination the disc and hypocotyl become a 
deep purple-brown in colour. Cotyledons not withdrawn from the endosperm on 
germination. 

L. miraculosus Miq. was described from specimens collected at Wicklow and 
York, by Preiss in 1840-41. 

I have not seen the type, but have been informed hj Professor Le Comte 
that only fragments of it remain in the Paris Herbarium. I have also been 
advised by Professor Carl Lindman that it is not represented at Stockholm, and 
that "they do not possess by far all the species of Preiss' collection." 

I have examined specimens from practically the type locality which answer 
in nearly everj'' detail to the original description. 

The leaves figured by Ettingshausen (Uber die Blattskelette der Loranthaceen, 
tab. ii.. Fig. 27, 29) match specimens from Dongarra, Mogumba, Cunderdin and 
other localities in Western Australia. 

Bentham, (Flora Australiensis, iii., 394) unites L. miraculosus Miq., and L. 
Melaleucae Lehm., and reduces them to a variety of L. pendulus Sieber, var. 
parviflorus Benth. 

It appears to me that he also includes L. Gaudichaudi DC. by including 
WooUs's PaiTamatta specimens under his var. parviflorus. His description of 
var. parviflorus is rather vague and would apply to either L. Gaudichaudi or L. 
Melaleucae. "Leaves small and narrow; flowers small, often 4-merous, the central 
ones sessile." 

I have only seen one specimen referred to by Bentham under var. parviflorm, 
that of C. Wilhelmi, from Port Lincoln, which undoubtedly answers to the des- 
cription of L. Melaleucae. 

Other localities quoted by Bentham are New England, Barrier Range, in- 
terior of South Australia, and Swan River. These in my opinion, are referable 
to L. miraculosus Miq., and Rottnest Island to L. Melaleucae Lehm. The? latter 
appears to be a narrow-leaved form of the former. 

Professor Ewart, commenting upon this species (Proc. Roy. Soe. Yict., N.S., 



134 THE LORAI^THACEAE OF AUSTRALIA^ iv., 

xxxi., 1918 (1919), 374), remarks, "Although the differences from L. pendulus 
are not very great (smaller leaves and flowers and the latter often 4-partite, etc.), 
the plant seems to form a constant and well marked type, Avith a facies distinct 
from L. pendulus. 

He then proceeds to discuss the position of L. miraculosus in conjunction 
with L. Melaleucae. Miquel (Plantae Preissianae, i., 281) quotes a species, No. 
6, under a manuscript name of Lehmann's, as L. Melaleucae, which is the same 
as the following species. No. 7, L. miraculosus, and strict priority rule requires 
that the earliest name should stand. 

Lehmann, as editor of the Avork, presumably consented to the publication of 
the manuscript name, but as the Avhole article on Loranthaceae is by Miquel, it 
would seem to be more apjDropriate to retain his name for the species, namely L. 
miraculosus Miq. 

Synonym. — L. pendulus Sieb. var. ■parviflonis Bentli. 

Range. — This is a variable species, particularly in the leaves and flowers and 
also to a lesser extent in the cymes. The typical form extends from Western 
Australia to New South Wales, as Avill be seen by the folloAving localities: — 
Western Australia: Murchison River (on Eucalyptus gomphocephala and Fusanus 
acuminatus). Greenough Dist. (J. Drummond, Hook. Journ. Bot., 5, 1853, 143), 
Geraldton, Dongarra (on -Acacia rostelUfera), Hills in the vicinity of Irwin and 
Dongarra (Diels and Pritzel, Bot. Jahr., 35, 1905, 176), Mogumber (on Fusanus 
acuminatus), Swan River (quoted by Bentham), York (The type locality, on 
Eucalyptus gomplwcephala) , Cunderdin (on Fusanus acuminatus), Bruce Rock, 
Merreden District (closely approaching the variety Boormani) , near Mt. Malcolm 
(■'Parasitic on the Quandang, Fusanus persicarius and homoplastic with it." S. 
L. Moore, Journ. Bot., xxxv., 1897, 169). South Australia: Ooldea (recorded as 
L. pendulus in Trans. Roy. Soc. S.A., xli., 1917, 379. The flowers are highly 
coloured and resemble the flowers of L. MiqueUi in a dry state, and Avere it not 
for the central one being sessile, and the short leaves, the specimen would pass 
for that species. Mr. Black in the same Journal, xliii., 1919, 29, records it under 
L. miraculosus : "Leaves 1-6 cm. long, thick, nerveless, oblanceolate ; the central 
flower of each partial cyme is sessile; corolla 15-20 mm. long.") Interior of 
South Australia (quoted by Bentham under L. pendulus var. parviflorus) , Robe, 
Yorke Peninsula (on Melaleuca parviflora (recorded in Trans. Roy. Soe. S.A., 
xHii., 1918, 29), East Plains (The specimen is not typical, but is inclined towards 
the variety Boormani in the shape and length of the leaves). There is a specimen 
in the National Herbarium, Sydney, from South Australia (W. Gill), without 
definite locality, which is nearly typical L. miraculosus. Victoria: Dumosa (on 
Loranthus linophyllus, the latter on Casuarina Luehmanni), Wycheproof, River 
Murray, on Santalum laficeolatus (Ned. Kruidk. Arch., iv., 1856, 105), Genoa 
River, on Angophora intermedia (Vict. Nat., xxxvi., 1920, 152). 

New South Wales: Mt. Bolton, near Moruya (on Casuarina stricta). South 
West Plains below Narrandera, (on Loranthus MiqueUi, the latter on Eucalyptus 
hemiphloia, var. microearpa), Albury Road, near Wagga Wagga, (on L. MiqueUi 
Lehm., the latter on Eucalyptus dealbata. {L. miraculosus Avas in flower and L. 
MiqueUi in fruit), BaiTier Ranges (quoted by Bentham under L. pendulus Sieb. 
var. parviflorus), Zara, Wanganella (on Fusanus acuminatus, Eremophila longi- 
folia, Acacia pendmla, and Myoporum acuminatum), Temora (on Fusanus acumi- 
nata), Barmedman (on Fusanus acurmnatus), Parkes (on Geijera parviflora), 
Harvey Ranges (showing example of double parasitism on L. MiqueUi, tlie latter 



BY W, F. BLAKELY. 135 

on Eucalyptus conica), Wellington (on Santalum and, in three instances, on L. 
pendulus, the latter in those three instances growing on Eucalyptus hemipMoia) , 
Dubbo (on Pittosporum phyllyraeoides. Another specimen showed self parasi- 
tism), Cobar District ("Small pretty flower, on the Wilga trees as a rule; flowers 
in June," These Proceedings, xxxvi., 1911, 523), Moonbi, New England, Narrabri 
(on EremopMla Mitchelli), Pilliga East State Forest, Co. Baradine, 15 miles E. 
of Wangan (on Fusanus aciiminatus). 

Affinities. — This species appears to have much in common with L. congener 
Sieb. in general appearance, and in morphological characters; it is, however, more 
glaucous and glabrous, and the leaves are more variable, with a relatively different 
venation, the cymes and flo^vers are also more glabrous, and the fruits are yellow, 
not green when ripe. 

Its affinity with L. Gaudicliaudi DC. lies chiefly in the similarity of the leaves 
with those of the variety Melaleueae. It is, however, sharply separated in the 
shape of the fruits, which are oblong or elliptical, while those of L. GaudicJiaudi 
are globose and bright -red, or deep pink in colour. 

J. Drummond (Hooker's Journ. Bot., v., 1853, 143) refers to "L. miraculosus 
from Greenough District," and compares it with "L. Quandang Lindl., in foliage 
of which there is a close resemblance in shape and size, but the former is always 
glabrous, and the latter pubescent." Drummond's observations are both interest- 
ing and correct. 

Fodder Value. — Spencer Le M. Moore, writing on the Camel food of Western 
Australia (Lon. Journ. Bot., ssxv,, 1897, 169) records this species under L. 
pendulus var. parviflorus Bth. as an excellent fodder plant. 

Mr. W. Bauerleu writing from Wellington, N.S.W., 11.2.1903, states that 
cattle eat the drooping shoots as high up as they can reach, while they leave the 
host plant, L. pendulus, alone even when far easier to reach. He also observed 
the cattle eating it and the var. Boormani, at Mt. Hope, near Cobar, in 1904. 

Hosts. — Casuarineae : Casuarina Luehmanni Baker. Loranthaceae : Loranthus 
linopTiyllus Fenzl., L. MiquelH Lehm., L. pendulus Sieber. Santalaceae: Fusanus 
acuminatus R.Br., F. persicarius F.v.M., Santalum lanceolatiis R.Br. Pittospora- 
ceae : Pittosporum phyllyraeoides DC. Rutaceae : Geijera parviflora Lindl. 
Leguminosae : Acacia rosteUifera Benth., A., ijendula Cunn. Myrtaeeae : Eucalyp- 
tus Baueriana Schau., E. conica Deane & Maiden, E. gomphocephala DC, Ango- 
phora intermedia DC., Melaleuca parviflora Lindl. Myoporaeeae: EremopMla 
longifolia R.Br., E. Mitchelli, Myoporum acuminatum R.Br. 

{a) Var. Melaleucae (Tate), n.comb. (Plate vi.) 

Tate, Trans. Roy. Soc, S.A., iii., 1879-80, 68 (as L. pendulus Sieb, var. 
Melaleucae; Lehm., Plant. Preiss, i, 1844-45, 281; Walp., Kept. Syst., v., 1845- 
46, 938; Miq., Ned. Kruidk. Arch., iv., 1856, 105; Benth., B.FL, iii., 1866; Et- 
tings, Uber die Blatts, der Lor., 1871, p. 11, Tab. ii., Figs. 13, 14. 

The following is a translation of the original description : — Glabrous, branches 
and branclilets terete; leaves opposite or subopposite, tapering into a somewhat 
long petiole, coriaceous, narrow spathulate, blunt, gi'adually contracted into the 
petiole, veinless or obscurely one-nerved, or usually somewhat veiny or reticulate 
above. Flowers .corymbose, pedicellate, 4-merous, supported at the base by a 
glabrous, oblong, concave bract. Calyx conical or cylindrical, the limb some- 
what long, spreading truncate: On the branches of Melaleuca, Rottnest Island, 



136 THE LORANTHACEAE OF AUSTEALIA, iv., 

18th August, 1839, (Herb.) Preiss No. 1616. A small shrub with numerous 
forked branches, the old branches cylindrical, covered with a smooth dark grey 
bark; leaf sears tuberculate; branchlets subterete. Leaves 2-4 cm. long, 6-8 mm. 
broad at the top, easily detached; petioles semiterete, confluent with the lamina, 
i to nearly 1 cm. long. Pedicels rounded, thickened upwards, 6 mm. long. Calyx 
tube 7 mm. long, carnose, the limb very long, dilated, somewhat undulate, trun- 
cate, scarcely repand. 

The following is a translation of Ettingshausen's description of the leaves: — 
Leaves oblong, wedge-shaped, petiolate, apex obtuse; venation uniting; primary 
nerve distinct, branching below the apex; secondary nerves 1-3, thin on both 
sides above the base, directed towards the margin; tertiary nervesi small, scarcely 
conspicuous. 

The type is in the Paris Herbarium, but like L. miraculosus, it is but a mere 
fragment. 

Mueller (Ned. Kruidk. Arch., iv., 1856, 105) expressed the opinion that it 
was the same as L. miraculosus Miq. ; I think however it is sufficiently distinct to 
warrant a varietal name until we know more about it. 

There is a specimen in the National Herbarium, Sydney, collected by C. 
Wilhelmi at Port Adelaide, South Australia. It does not differ in any way from 
Wilhelmi's Port Lincoln specimen. 

Professor R. Tate (Trans. Roy. Soc. S.A., iii., 1879-80, 68) recorded it as 
L. pendulus Sieb. var. Melaleucae without a description. His specimen which 
was kindly lent by Professor Osborn of the University of Adelaide Herbarium is 
somewhat larger than Wilhelmi's specimen, being intermediate between the typical 
L. miraculosus and vai\ Boormani. 

This variety is a smaller and more compact shrub than the typical L. mira- 
culosus, with usually narrow cuneate or narrow spathulate leaves, the narrow 1- 
nerved, and the larger or broader ones 1 — 3-nerved. Buds variable, sometimes 
quite normal, but in some specimens they are longer and more robust and pass 
imperceptibly into var. Boormani. While it keeps to the coast, the leaves are 
fairly uniform, but, as it extends its range to the drier interior, they become 
gradually larger until the variation is excessive, hence the description of a new 
variety, var. Boormani. Nevertheless, careful field observations are necessary to 
determine the exact relationship between it and L. miraculosus Miq. 

It is singular that it is invariably parasitic on the genus Melaleuca. 

Synonyms. — L. Melaleucae Lehm., L. pendulus Sieb. var. parviflorus Benth., 
L. pendulus Sieb. var. Melaleucae Tate. 

Bange. — Western Australia: Rottnest Island (A. Cunningham, Preiss, No. 
1616) and probably Swan River (quoted by Bentham). South Australia: Bunda 
Plateau (on Fusanus). Recorded by Professor Tate in Trans. Roy. Soc. S.A., 
ii., 122 (1878-9), as "L. pendulus Sieb. var. Melaleucae Lehm." Specimen al- 
ready referred to. Fowlers Bay (on "tea-tree"), Murat and Denial Bays (on 
Melaleuca cymhifolia Bentli. (recorded by J. H. Maiden, Trans. Roy. Soc. S.A., 
xxxii., 1908, 263) ; Unley Scrub (in Herbarium, University of Adelaide), Compass, 
No. 1025, on "tea-tree," probably collected by Professor Tate (in Adelaide Her- 
barium), Port Lincoln (in Herb., Melbourne), Andi'ossa (on Melaleuca ericifolia 
in Adelaide Herbarium), also recorded by 0. Tepper, Trans. Roy. Soc. S.A., iii., 
1879-80, 39. South Yorke Peninsula (in Adelaide Herbarium), Port Vincent, 
Yorke Peninsula (Flowers scarlet, petals free, 13-14 mm. long; Flowers corym- 
bose, 3 rays, bearing 3 flowers each; leaves opposite, cylindrical, but .slightly 



BY W. r, BLAKELY. 137 

verticillate, comparatively obtuse, sulcate; on Melaleuca Preissiana. This is the 
narrowest leaved specimen I have seen of this variety), White Lagoon, Kangaroo 
Island (on Melaleuca sp., vide J. H. Maiden, Proc. Linn. Soc. N.S.W., xxix., 1904, 
696), near Port Adelaide, Murray River (Ned. Kruidk. Arch, iv., 1856, 105). 

Affinity. — This variety bears a striking resemblance to L. Gaudichaudi DC 
in the leaves, when they are one-nerved; but it will be observed that the young 
shoot and leaves are strictly glabrous, whereas the leaves of L. Gaudichaudi are 
minutely tomentose in the early stages. 

The buds of var. Melaleucas are larger and more striate and of a yellowish 
colour. 

The fruits of both, when available, afford a ready means of discrimination be- 
tween them; they are yellowish, and slightly urceolate or elliptical in var. Mela- 
leueae, and bright pink, or light red, and constantly spherical in L. Gaudichaudi. 

ITosfs.— Santalaceae : Fusanus sp. Myrtaceae: Melaleuca cymbifolia Bth., 
M. ericifolia Sm., M. Preissiana Schau. 

(6). Var. BoORMANi^ n.var. (Plate vii.) 

Frutex glaber plerumque pendiilus ramis tenuibus nonnumquam ultra 3 pedes 
saepe caespitem densum generans fusiformis. Folia angusta late linearia vel 
lineari-oblonga 2-10 em. x 3-10 mm. crassa plerumque unicostata. Cymi saepe 
tri-radiati. Pedunculus communis tenuis 2-3.5 cm. Pedicella gemmaeque lon- 
guires quam in typiea. Fructus ovatus ad cylindraeeus 7-9 mm. long, 5 mm. diam. 

Plants glabrous, usually pendulous, with slender branches, sometimes ex- 
ceeding 3 feet in length and often forming dense masses. Union attachment 
more elongated than the typical form. Leaves narrow to broad linear or linear 
oblong, 2-10 cm. by 3-10 mm., thick, usually one-nerved, or some of the broader 
leaves faintly 3-nerved. Cymes frequently 3-branched, the common peduncle 
slender, 2-34 cm. long; pedicels and flowers longer than the typical form, other- 
wise the same. Fruit cylindrical-ovate. Amber Yellow (Fig. 28, No. 1-4), 
changing to Straw Yellow (Fig. 31, No. 1-4, Rep. de Coul., Dauthenay), 7-9 mna. 
long, 5 mm. in diameter, crowned with the persistent calyx limb. Seeds obconie 
to elliptic, with a small spongy base, indistinctly 5— 6-furrowed. Viscin white, 
copious, very sweet. Endosperm at first white, but on the germination of the 
embryo turns to a dark green; hypocotyl dark green, terete and vermeose, some- 
times attaining 10 mm. in length. Embryonic cotyledons narrow oblong, slightly 
broader at the top, 3 mm. long. 

This form appears to be more widely distributed than the normal form. It 
is very common throughout the western and north-western parts of New South 
Wales, and is also found in Queensland and South Australia. 

This variety is relished by stock. 

Bange. — South Australia: Mt. Lyndhurst and Ediowie (in Adelaide Her- 
barium); Victoria: Lake Hnidmarsh (on L. Miquelii, the latter on Eucalyptus 
bicolor) ; New South Wales: Broken Hill (on Acacia cana), Nulty-Toorale, Paroo 
R., Paldrumatta Bore, Wilcannia (on Eremophila Stuartii), Bourke (on Myo- 
porum platycarpum, Ventilago mminalis, Geijera parviflora), Shuttleton (on 
Geijera parviflora), Cobar (numerous hosts). Mount Hope (on L. Miquelii, the 
latter on Eucalyptus hemiphloia), Gilgunnia (on G. parviflora), Wjttagong, Car- 
gelligo-Condobolin, West Bogan (on Fusanus acuminatum), Coolabah (on G. 
parviflora), between Boggabri and NaiTabri (on L. linophyllus, the latter on 
Cas2iarina lepidophloia) , Narrabri West (on L. Miquelii, the latter on Eucalyptus 



138 THE LORANTHACEAE OF AUSTRALIA^ iv., 

popidifolia) , Narrabri. Queensland: Diamantina R. (in Queensland Herbarium), 
Bulloo R. (on Fusanm acuminatus), Aramae, Muttaburra (on Santalum lanceo- 
latum), Jericho. 

Diifering from L. miraculosus Miq. in tlie longer and narrower leaves and in 
the longer peduncles, pedicels and buds. 

It is a larger, coarser, and more pendulous plant than var. Melaleucae, with 
relatively longer leaves and buds. 

Named in honour of John Luke Boorman, Botanical Collector, Botanic 
Gardens, Sydney. 

Hosts. — Casuarineae: Casuarina Luehmcwni B.. T. Baker. Santalaceae: Fusa- 
niis acuminaUis R.Br., Santalum lanceolatum R.Br. Loranthaceae : Loranthus 
Miquelii Lehm., L. linophyllus Fenzl. Leguminosae : Acacia aneura F.v.M., A. 
cana Maiden, A. excelsa Benth., A. harpophylla Cunn., A. homalophylla A. Cunn. 
Rutaceae: FUndersia maculosa F.v.M., Geijera parviflora Lindl. Rhamnaceae: 
Ventilago viminalis Hook. Myrtaceae: Eucalyptus bicolor A. Cunn., E. liemi- 
phloia F.v.M; Myoporaeeae: Eremophila Mitchelli Bth., E. Stuartii R.Br., Myo- 
porum platycarpum R.Br. 

(c). Var. PUBiGERA, n.var. (Plate vii., fig-s. 8, 9.) 

Surculi gemmaeque pilum rufum praebint. Folia angusta ad lato-laneeolata 
subacuta 3-costata in brevem teretem petiolum a,ttenuata. Cymi bi-radiati. Gem- 
mae graeiles. Bracteae calycesque timbriati pilis mollibus. Basis calycis dense 
einereo-tomentosa. Stylum tenuis apiee fere eontorto. Stigma capitatum. 

Leaves naaTow to broad lanceolate usvially, somewhat acute, 3-nerved, 1-3 
inches long, tapering into a short terete petiole. Young shoots and buds minutely 
hoary or partly rufus pubescent. Cymes 2-branehed, buds slender. Bracts and 
calyces fiinged with a rather conspicuous border of soft hairs, base of calyx 
densely hoary. Style very slender, almost spirally twisted at the top, stigma 
capitate. 

Differing from var. Boormani in the vestiture of the buds and nascent parts, 
and, like it, the peduncles and pedicels are not carnose as in the typical L. mira- 
culosus. 

The style is more twisted at the top' than the preceding, and it is also longer, 
and the stigma is larger. 

Loc. — 35 miles west of Cobar (on Eucalyptus sp. and Acacia homalophylla ^ 
L. Abraham, No. 801, 6, 1912). 

Hosts. — Leguminosae: Acacia homalophylla A. Cunn. Myrtaceae: Eucalyptus 
sp. 

13. LORAXTHUS Gaudichaudi DC. (Plate viii.) 

Prod. Syst. Yeg., iv., 1830, 295, non Hugel; Benth., B.FL, iii., 1866, 594, 
as L. pendulus Sieb., var. parviflorus (in pari) ; Etting., Uber die Blatts. der Lor., 
1877, Taf. ii., Fig. 15. 

The following is a translation of the original description by G. Don in Gen. 
Hist., iii., 419, as Dendropldhoe Gaudichaudi: glabrous; branches terete; leaves 
opposite, linear, obtuse, thickish, attenuated at the base; peduncles axillary, much 
shorter tlian the leaves, divaricately 1)ifid at the apex, 2 — 4-flowered; parts of 
flowers equal; Howers tetrandrous; petals spathulate; anthers ovate, inserted by 
the base, H.P.G. Native of New Holland, where it was collected by Gaudichaud. 



BY W. F. BLAKELY. 139 

Lorantlius GaudicJiaudi, DC. prod, -i p.. 295. Leaves 10-12 lines long', and 2 lines 
broad. Flowers 5-6 lines long. Style filiform." 

The following is a more detailed description : — Small compact shrubs with a 
ball-like union, forming' globular masses, 6-36 inches in diameter, usually forming 
on the ends of the branches of the host. Young branches and leaves infested 
with a minute ferruginous deciduous vestiture; old branches lenticular-scabrous, 
and marked by the prominent scars of the petioles. Adult leaves glabrous, op- 
posite, or occasionally a few alternate, linear-cuneate to spathulate or oblong- 
lanceolate, obtuse, one-nerved with very obscure lateral veins, tapering into a 
semiterete, 3-5 mm. petiole, usuallj^ 2-3 cm. long, sometimes exceeding 4 cm., the 
mai'gins very thin. Cymes slender, chiefly axillary, usually 2 — 3-branehed, each 
branch bearing 3 flowers, the central flower of each triad sessile, the lateral one 
on shoi't slender pedicels. Flowers crimson-purple, smaller than L. miraculosiis ; 
buds ciavate, minutely scurfy as well as the floral branches, 10 mm. long, very 
slender; bracts small, navicular, ciliate; segments 4-5, free, linear, crimson. Fila- 
ments and style the same colour as the segments. Anthers adnate, ovate, 1-2 mm. 
long. Stjde slender; stigma capitate, rather large. Fruits globose bi'ight red, 
3-4 mm. in diameter. Seeds elliptical, 2 mm. long, the base somewhat spongy, 
indistinctly 4-furrow€d. Endosperm white, except the base which appears to be 
constantly pinkish. Hypocotyl gTeen, lengthening out to about 5 mm., with a few 
prominent tubercles. Disc somewhat conical. Embrj^onic cotyledons linear ob- 
.tuse, to semiterete, not Avithdrawn on germination. Primary leaves linear, obtuse, 
thick, 2-4 mm. long. 

It is not quite clear where Gaudichaiid collected the plant which bears his 
name, as he collected plants on the West Coast, Port Jackson, Botany Bay, and 
the Blue Mountains (vide F. M. Bailey, Proe. Roy. Soe. Q'land, viii., 1889-90, 22) . 

It is most likely that the specimens were collected in the Port Jackson dis- 
trict, as the species is common along the Parramatta River, in the vicinity of 
Parramatta. Most of these places were easy of access to the early botanical 
collectors. 

So far, I have not seen specimens of this species from Western Australia. 
The L. Gaudichaudi of Hugel (Plantae Novae HoUandiae, 1848, 56) is, in my 
opinion, a form of L. linophyllus Fenzl, as the description is applicable to it. 

The leaves of L. Gaudichaudi DC. figured by Ettiiigshausen (Uber Die 
Blattskelette der Loranthaceen, Taf. ii., Fig. 15) cannot be separated from many 
specimens both in shape and venation over a large range of specimens from Port 
Jackson and other districts. 

Bentham does not refer to L. Gauddchaudi DC. in the '"Flora"; he evidently 
overlooked it, as it is referred to by Mueller in his Report of the Burdekin Ex- 
pedition (1860), which is quoted by Bentham in the "Flora." 

It appears to me that Woolls' specimens from Parramatta, quoted by Bent- 
ham under L. pendulus Sieb. var. parviflorus, are referable to L. Gaudichaudi 
DC. or L. congener Sieb., as both species are plentiful in the above locality. It 
is quite possible that he also regarded the Parramatta plant as L. miraeulosus 
Miq. or L. Melaleucae Lehm. 

Had he ample material of these species, especially the fruits, I am confident 
that his decision would have been different. 

Mueller (Report Burdekin Expedition) refers to this species as follows: "An 
extreme form seems the L. Gaud/ichaudii, varying with 4 and 5 petals, having the 
leaves and flowers reduced to remarkable smallness, and the anthers by diminu- 
tion of length altered to an ovate form." 



140 THE LORANTHAGEAE OF AUSTRALIA^ iv., 

The only other record I am able to trace of this species in Australian litera- 
ture is by WooUs (A contribution to the Flora of Australia, Botany of Ash 
Island, 1862, 187). He says "L. Gaudichaudi DC grows principally on Mela- 
leuca." 

Synonyms.- — L. pendulus Sieb. var. parviflorus Benth. (in part), Dendroph- 
thoe Gaudichaudi G. Don. 

Bange. — This species is so far confined to New South Wales, and appears to 
be restricted to the coastal range, extending from Thirlmere, South of Sydney, 
to the Queensland border. Its appearance can be looked for in the northern 
State. 

Neiv South Wales: Thirlmere to Granville, Parramatta to St. Mary's and 
Richmond, Homebush to Rhodes, Lakemba,, Penshurst, Dorrigo Forest Reserve, 
Acacia Creek, Ash Is., Clarence River, 

Affinities. — Its closest affinity is L. miraculosus Miq. var. Melaleucae, from 
which it can be distingTiished by the minutely pubescent young parts, buds, and 
smaller flowers with their corresponding frailer and smaller peduncle and pedicels, 
also in the small globose fruit. 

From L. miraculosus Miq., it is distinguished by the preceding characters 
and in the smaller, one-nerved, narrow, cuneate leaves. The bright red fruits of 
L. Gaudichaudi bear some resemblance to the fruits of L. Preissii, L. Cambagei, 
and L. linophyllus ; the latter three are terete-leaved species. 

Phrygilanthus Bidtvillii would easily pass for this species when in fruit, as 
the foliage of some forms and the fruits are remarkably homoplastic, but the 
bifurcate peduncle, if present, is a distinguishing character. The anthers of 
coui'se are different in both. 

The common host-plant of L. Gaudichaudi, according to Mr. E. Cheel who 
has revised the genus Melaleuca, is M. genistifolia Sm., and not M. parvifoUa 
Lindly., as generally accepted. 

In the George's River district, this species forms dense compact masses, 
sometimes completely smothering the host to within a foot of the ground. It is 
in easy reach of cattle, but they do not appear to eat it. 

Hosts. — Loranthaeeae : Loranthus pendulus Sieb., L. congener Sieb. Myr- 
taeeae: Melaleuca genistifolia Sm., M. styphelioides Sm. 

14. LoRAXTHUs Preissii Miq. (Plate ix.) 

Lehm., Plant. Preiss., 1844-5, 280; Walp., Rept. Bot. Syst, v., 1845-46, 938; 
]\Iiq., Ned. Ki-uidk. Arch., iv., 1856, 105; Benth., B.FL, iii., 393 in^ part; forma 
(a) ; Wild., PI. Nov. Then., ii., 1909, PL 76; Engl, Engl, et Prantl, Pflanzenfam. 
as L. linophyllus Fenzl. 

The following is a translation of the original description: — Glabrous, 
branches terete, leaves subopposite terete, filiform, somewhat straight, subacute or 
blunt, i\(iiihy or succulent. Flowers on axillary two or three branched cymes 
shorter than the leaves, pedicellate, or the central one sessile; calyx shortly ob- 
conic, tlie limb uneven. Petals 5, hnear-spathulate, concave at the apex; stamens 
inserted on the upper quarter of the petals, of equal length and a little shorter 
than the filiform style which is thickened upwards. 

Parasitic on Acacia sp. near Perth, flowering, 8.3.18 Herb. Preiss. No. 1611. 

Similar to the preceding (L. Casuarinae), but the inflorescence and flowers 
ver>' different in shape. Branches and branchlets terete, straight, alternately 
forked, glabrous, smooth, dark green. Leaves straight or curved, 4-5 cm. long, 



BY W. F. BLAKELY. 141 

somewhat terete, wrinkled and somewhat rigid when dry, bluntish or acute, 1-2 mm. 
thick. Flowers cymose, the partial cyme two- or three-branched, supported on 
pedicels about 4 cm. long, the central flowers sessile; bracts flesh-coloured, ovate, 
acute, navicular, glabrous, or the apex somewhat ciliolate or torn, dark brown in 
a dry state; the peduncles 2i, cm. long. Buds cylindrical-elavate, a little inflated 
at the base, somewhat straight, the apex thick and obtuse. Calyx shortly ob- 
conic, somewhat cupular, 2 mm. long, the limb truncate or uneven. Petals 5, 
linear, the apex elliptical, dilated, somewhat acute, concave; stamens inserted a 
little below the expanded portion; anthers linear-spathulate, acute, 3 mm. long. 
Style filiform, slightly exceeding the petals, thicker towards the base; stigma 
scarcely capitate, somewhat truncate. 

Additional notes. — Glabrous compact shrubs, 1-3 feet in diameter; union ball- 
like; branches pendulous, terete, 1-4 feet long; leaves opposite, alternate or in 
clusters of 3 or 4 on the same specimen, terete or somewhat compressed, mostly 
slender, straight or curved, acute, IJ to 4 inches long. Inflorescence cymose, 
glabrous; cymes 1-3 together on the nodes of the old and young branches; 
peduncles slender, J to 1 inch long, considerably thicker at the base, usually only 
2 — 3-branched, each with a partial cyme of 3 flowers, the central flower of each 
triad sessile, the lateral flowers on slender pedicels 5-9 mm. long; bracts concave, 
navicular, ciliate on the margin, about as long as the calyx, but sometimes exceed- 
ing it; calyx cupular, the limb minutely denticulate-ciliate, and often undulate. 

Buds robust, reddish, 1 inch long, clavate at the top, angular or scarcely 
terete towards the base, the latter abruptly inflated to a diameter gTeater than the 
calyx, causing the rim of the calyx to bulge over the solid portion of the calyx. 
Petals 5-6, yellowish in the upper portion, reddish towards the base, narrow- 
spathulate to lanceolate from the insertion of the fllaments upwards, very narrow 
to within 4-5 mm. of the base, then abruptly broader than the top'. Filaments 
dark red, the adnate portion compressed and enlarged near the base, causing the 
corolla to become abnormally inflated at its junction with the calyx, the free por- 
tion about as long as the anthers. Anthers oblong, 3 mm. long; pollen grains 
bright yellow. Style angular, geniculate about 3 mm. below the stigma, and 
gradually narrowed upwards; stigma flat or repand. Fruit glabrous, globose, 
white, spotted red or flesh-coloured, often changing to a dark red and, according 
to Mueller, about the size of a pea (vide Trans. Vic. Institute, 1854-55, 128). 

Disc fleshy, raised around the base of the style., obscurely 5-angled, and con- 
spicuous on the ripe fruit. Seeds narrow elliptical, faintly 5-ribbed, 4-5 mm. 
long. Endosperm white, bony when dry; hypocotyl terete, verrucose; disc conical. 
Embryonic cotyledons narrow, spathulate, thick, 3 mm. long, not withdrawn from 
the endosperm on germination. 

Synonyms. — L. Preissii Miq., varietas ? Miq. (I.e.). I have not seen Miquel's 
specimen but, judging from the description, it appears not to vary to any gTeat 
extent from the typical form, which is fairly uniform, and in a large series of 
specimens the leaves are alternate on some branches and opposite on others on 
the same specimen. The flowers are also much brighter when exposed, than those 
eoUeeted in shady places. So far I have not seen a specimen differing materially 
in any characters from the typical form from Western Australia. L. scoparius 
Miq. (I.e.) I am inclined to regard as a synonym of L. Preissii Miq. It appears 
to have been described from imperfect material. Mueller (Rept. Burd. Exped.), 
when discussing L. scoparius ^ states that it is the same as L. Preissii. 

Range. — The South Australian localities quoted by Bentham {I.e.) are in all 
probability referable to L. Preissii Miq., as all the authenticated specimens ex- 



142 THE LORANTHACEAE OF AUSTRALIA^ iv.; 

amined by me from S.A. proved to be this species, and, strange to say, I have not 
seen a single specimen of the typical L. Unophylliis Fenzl from South Australia. 

Western Aiistralia: Avon and Irwin Districts (frequently on Acacias)^ York 
(on Acacia acuminata) , Gin Gin (on A. cyanophylla), Minginew (on Acacia sp.), 
Ernest Giles Range (on Acacia aneiira; also recorded in Trans. Roy. Soc. S. 
Aust., xi., 1892, 360 as L. linopJiyllus) , Strelly R. (on Acacia colletioides) . South 
Australia: Upper Coolanoorina Creek (on Acacia aneura; recorded as L. lino- 
phyllus), Sandhills, Lake Hart (on A. aneura), L. Eyre, Cooper's Creek, Lake 
Australia (on A. melanoxylon) , Dublin Scrub, Mypunga, Lyndhurst, Leigh's 
Creek, Tallaeallana Creek (on A. stenophylla) , Bunda Plateau, Crystal Brook, 
Roberts Town, Clarendon (on A. melanoxylon and on L. pendulus), Androssa, 
Barossa and Mt. Lofty Range (on A. melanoxylon and A. retinodes) , Murray R. 
to Spencer's Gulf, Overland Corner (on ''Needle Acacia"). Victoria: Toronto 
(on ^1. retinodes), Hawkesdale, Grampians (on A. deaurrens). Neiv South Wales: 
Darling R. near Wentworth (on A. tetragonophylla) , Broken Hill (on A. aneura, 
Myoporum platycarpum, A. tetragonophylla and A. aneura), Cobar (on Acacia 
sp.), Gunbar (on A. Oswaldi), Tibooburra (double parasitic with L. Exocarpi on 
A. tetragonophylla) , Queensland: North from Broken Hill, Diamantina. Northern 
Territory: Dalhousie Springs (on Acacia sp.) , Dalhousie Station (on A. salicina 
var. varians, recorded as L. linophyllus) , Idracowra, Finke R., G. of Carpentaria. 

It will be seen that L. Preissii is recorded for all the States of the mainland, 
and is found not only in the coastal districts of Western Australia, but is also 
plentiful in Central Australia, which is probably its home. The palatable fruits 
are eagerh' eaten by many birds which, no doubt, accounts for its wide range. 
Its flowering season appears to be from December to March, It is very singular 
that collectors have only noted this species growing upon Acacia, except in one in- 
stance, when it was found on Myoporum platycarpum. 

The three specimens from Victoria, the only ones I have seen from that State, 
\ary from those of the other States in the more slender foliage and slightly smaller 
flowers. The most robust and most handsome specimens are those from Central 
Australia, in which the petals are highly coloured. 

Affinities. — L. Preissii Miq. differs from L. linophyllus Fenzl in being strictly 
glabrous, both in foliage and flowers; the latter are more highly coloured, and the 
corolla is less angular and more inflated at the base; the fruits also are largei', 
glabrous and less contracted at the top than those of L. linophyllus Fenzl. 

L. Preissii Miq. is also closely allied to L. Camhagei, from which it differs in 
the shorter and glabrous leaves, larger and more robust glabrous buds; differently 
shaped calyx, and in the larger and more highly coloured fruits. 

According to Max Koch (Trans. Roy. Soc. S.A., xxii., Part ii., 1898, 101), 
the native name of this species is Partapee. 

Hosts. — Loranthaceae : Loranthus Exocarpi Behr., L. pendulus Sieber. Legu- 
minosae: Acacia acuminata Benth., A. aneura F.v.M., A. colletioides A. Cunn., 
.1. cyanophylla Lindl., A. decurrens Willd., A. Lehmanniana F.v.M., A. melanoxylon 
R.Br., A. Oswaldi F.v.M., A. retinodes Schlecht., A. salicina Lindl. var. varians 
Benth., A. sentis F.v.M., A. stenophylla A. Cunn., A. tetragonophylla F.v.M. 
^fyoporaceae : Myoporum platycarpum R.Br. 

Var. DiDYMA, n.var. (Plate ix., fig. 9.) 

Glaber, ramis teretibus; foliis teretibus; pedunculate 2-floribus, 5 mm. longo; 
pedieellis graeilibus leviter compressis 0-8 mm. longis; alabastris formae typicae 
pimilibus, 1.3 mm. longis. Calyce obconico limbo repando; fmetum non vidi. 



BY W. r. BLAKELY. 143 

Flowers in pairs on slender, slightly compressed pedicels, G-8 nnn. long; the 
common peduncle 5 mm. long. Buds similar to the typical form, 13 mm. long. 
Calyx in the spent flowers conical. Fruit not seen. Queensland, Diamantina, Dr. 
T. L. Bancroft, April 1892, No. 1078. 

The only specimen seen of this variety is somewhat fragmentary, but the 
didymous inflorescence is a striking contrast to the typical triflorous cymose in- 
florescence of this series, and is apparently constant; when in bud, it is some- 
what similar to L. Mitchelliana, but the flowers are not tubular, the segments are 
sepai-able to the base. The common peduncle appears to be articulate in the 
middle; if this is so, it is unusual in this series. On the other hand it may be 
bifurcate in perfect specimens. It gives the impression of a secondary peduncle, 
but from the scant material it is difficult to come to a definite conclusion. Further 
field observations are necessary to define it clearly. 

15. LoRANTHUS Cambagei^ n.sp. (Plate x.) 

Quoted by Bentham (B.Fl., iii., 393) under L. Unophyllus Fenzl. as (&), "with 
slightly tomentose flowers from Brisbane to Port Jackson and Illawarra." 

After a careful comparison with L. lino2Jhyllus Fenzl, and taking into con- 
sideration its wide range, and the numerous specimens which came under my 
scrutiny, together with my fleld knowledge of the plant, I feel justified in raising- 
it to specific rank. 

Frutex plerumque densus cinereus pendulus giobosus ramis tenuibus glabris 
nisi minute tomentosis apicibus. Folia teretia gracillima 3-8 unc. longa. In- 
floreseentia cjToiosa cymis axillaribus aliquanto tomentosis 2-4 ramatis utroque 3 
floribus centrali sessili. Gemmae porriginosae graciles 1.5-2 cm. longae. Corolla 
petalis subtiUima, Filamenta basi non caleaxata. Antherae adnatae oblongae 2 
mm. longae. Fructus giobosus fere glaber eolorem carnis exhibens 3-5 mm. diam. 

Usually compact, j)endulous, ashy-grey shrubs with a ball-like union, and 
slender glabrous branches, except for the very young parts which are minutely 
tomentose, this passing away with age, 2-4 feet long, forming a mass 1-4 feet in 
diameter. Leaves very slender, the young ones slightly tomentose, glabrous when 
fully developed, opposite, alternate, occasionally in little clusters of 3 or 5, terete, 
wrinkled, 3-8 inches long, acute or with a small oblique mucro. Inflorescence 
ejonose; cymes axillary, occasionally terminal on very old plants; the common 
peduncle terete, 1^-2 cm. long, or longer, bearing an umbel of 2-4 rays, with a 
partial cyme of 3 flowers, the central one of each triad sessile, the lateral flowers 
on short pedicels. Bracts minutely pubescent, narrow, acute; calyx cupular, very 
small, tomentose at the base; the limb eiliate and almost glabrous. Buds slender, 
slightly tomentose, terete, clavate at the top, the base somewhat inflated, under 2 
em. long. Petals 5, free for their entire length, very thin, reddish to purple- 
brown inside, flesh-coloured or hoary pink outside, narrow-lanceolate, often with 
a central line. Filaments reddish, the free portion compressed, 5 mm. long; 
anthers adnate, narrow-linear, or linear-oblong, 2 mm. long. Fruit globose, 
crowned Vidtli the persistent calyx-limb into a verj^ short tube, frosted or minutely 
tomentose, pearl-white to flesh-colour, 3-4 nam. in diameter; disc circular, con- 
spicuous. Seed elliptical, with 5 scarcely prominent furrows around it, the base 
spongy; the elastic horns, corresponding with the 5 furrows, very thin, exceeding 
the seed by about 1 mm. Endosperm white, hard and bony when dry. Hypo- 
cotyl very slender, verrucose; sudoral disc flat. Embryonic cotyledons narrow- 



144 THE LORANTHACEAE OF AUSTRALIA^ IV., 

lineal', obtuse, not withdrawn on germination; primary leaves semi-terete, acicular. 

The description is drawn up from specimens from Grros6 River and banks 
of the Nepean, near its confluence with the Grose ; parasitic on Casuarina Cunning- 
hamiana Miq., J. H. Maiden and R. H. Cambage, Ko. 758, 9, 1906. Robert Brown 
collected at this spot about May, 1803, and January, 1805; his specimens are re- 
ferred to by Bentham (I.e.). 

Named in honor of Mr. R. H. Cambage, who for many years has taken a 
keen interest in the native flora, and has contributed many valuable specimens to 
the National Herbarium. 

Synonym.— L. linophyllus Fenzl var. parviflorus A. G. Hamilton (Proc. 
Linn. Soe. N.S.W., ii., Second Seriesj, 1887, p. 282) is probably the same as L. 
Cambagei, but as there is no description or specimen in existence, its identity 
cannot be confirmed. 

Range. — New South Wales: Burrinjuck and Taemas Bridge, south of Yass 
(on Casuarina Cunninghamiana) , Narrandera, Barmedman (on C. lepidopMoia) , 
Bowan Park, near Cudal (on C. Cunninghamia/na) , Trangie and Nyngan (on C. 
lepidopMoia), Nepean and Grose Rivers (on C. Cunninghamiana) , Newport to 
Barrenjoey (on C. glauca), Agnes Bank^, Windsor (on C. suherosa, Melaleuca 
linearifolius and a cultivated plum), Tuggerah Lakes (on C. glauca), West Mait- 
land, Belmont (on C. glauca), Tilligerry, near Booral, Mudgee and Dubbo (on 
C. sp.), Narromine (on C. Luehmanni), West Bogan (on Casuarina and Eremo- 
phila Mitchelli), Bugaldie (on C. Cunninghamiana), Warrumbungle Ranges, 20 
mis. N.E. of Boggabri (on C. Cunninghamiana) , Narrabri (on C. lepidopMoia), 
Yarrawin Station, Barwon River, Warialda (on C. lepidopMoia), Casino (on C. 
torulosa), Richmond R. (on C. glauca), Kyogle (on Eugenia Smdthii), WaUan- 
garra. Queensland: Brisbane R. (on Casuarina), Bulimba (on Notothixos 
suhaureus, the latter on C. glauca), Burpengary, Imbil (on C. Cunninghamiana) , 
Surat, Stannary Hills, Evelyn, Herberton (on C. torulosa). 

At Bowan Park, near Cudal, this is the most common Loranthus fre- 
quenting Casuarina Cunninghamiana, and occasionally parasitic on L. pendulum 
and L. Miquelii. It is strikingly constant in its characters, as no marked varia- 
tion was observed by me in the numerous plants throughout the district. The 
absence of L. linophyllus from this particular district is also very marked. In the 
fruiting season it is visited by many frugivorous birds which are particularly 
partial to its small juicy fruits. The following have been noted by me feeding 
upon its fruits: — Mistletoe-Bird, Dicaeum hirundinacetim {vide Leach, No. 336); 
Golden Rumped Diamond-Bird, Pardalote xanthopygius, No. 341; Silver-eye, 
Zosterops coerulescens, No. 334; Soldier-Bird, Myzantha garrula, No. 368; Leather- 
head, Tropidorhynchus corniculatus , No. 374; Rosella, Platycercus eximius, No. 
202; Cockatoo-Parrot, Calopsittacus novae-hoTlandiae, No. 196; Galah, Cacatura 
roseicapilla, No. 194. 

Affinities. — L. Cambagei is closely allied to L. linophyllus Fenzl, from which 
it differs as follows : — 

L. Cambagei, n.sp. L. linophyllus Fenzl. 

Branches glabrous, or the young Branches hoary-tomentose for two or 

shoots minutely pubescent, the pubes- more years, 
cenee passing away the first year. 

Leaves very slender, only the very Leaves hoary, thick and coarse, 

young on&s slightly tomentose, 2-8 sometimes becoming glabrous with age, 
inches long. 2-6 inches long. 

Cymes slightly tomentose or scurfy; Cymes densely woolly-tomentose ; the 

peduncles slender in proportion to the common peduncle short and stout, 

hoary-tomentose, li cm. long, usually 



BY W. F. BLAKELY. 145 

number of flowers, scantily pubescent, 1 cm.; lateral pedicels stout, 3-4 mm. 

rarely less than 2 cm. long; lateral long. Flowers few. 
pedicels slender, 5-7 mm. long. Flowers 
numerous. 

Buds slender, 1.5-2 cm. long. Petals Buds robust, 2-2.5 cm. long. Petals 

very thin. tliick. 

Filaments terminating without a Filaments ending m a spur-like cal- 

eouspicuous spur at the base, 5 mm. losity at the base, the upper free por- 

long; anthers 2 mm. long. tion 4 mm. long; anthers 3 mm. long. 

Fniit globose, flesh-coloured, 3-5 mm. Fruit ureeolate to elliptical, 5-7 mm. 

in diameter, nearly glabrous. in diameter, hoary-tomentose. 

From L. Preissii Mq., it is distinguished by its longer and more slender 
leaves ; pubescent young shoots ; slender mealy buds, and smaller f niits. The buds 
of L. Preissii Miq. are more inflated at the base than those of L. Camhagei. 

It is interesting to note that L. Camhagei has not yet been foimd associated 
with L. Preissii Miq., and the latter species has not yet reached the eastern and 
north-eastern coast. 

Hosts. — Casuarineae: Gasiiarina Cwnninghamiana Miq., C. glauca Sieb., G. 
lepidopMoia F.v.M., C. Lwehmanni R. T. Baker, C. suberosa Ott. et Diet., C. 
torulosa Ait. Loranthaceae : Loranthus pendulus Sieb., L. Miquelii Lehm., Noto- 
tMxos suhaureus Oliv. Rosaceae: Plum, var. Lutherboro. Myrtaceae: Melaleiica 
Unearifolius Sm., Eugenia Smithii Poir. 

16. LORAN^THUS LiNOPHYLLUS Fenzl. (Plate xi.) 

Hueg. Enum., 1837, 56; Walp., Rep. Bot. Syst., ii., 443; Huegel, PI. Nov. 
HoU., 189 ; Benth., B.Fl., iii., 393 (in part) ; Muell., Key. Viet. PI., i., 273 ; 
Moore and Betche, Fl. N.S.W., 228; Bail., Q'land. Fl., Part v., 1380. 

Bentham and Hooker (Genera Plantamm, iii., 208) credit Presl with being 
the author of L. linopJiyllus, not Fenzl, which appears to be a mistake. 

This is the forma (c) of Bentham {I.e.) ''with woolly-tomentose flowers, in 
New England and West Australia." 

I have not seen the original description, and, as Bentham's description in- 
cludes more than one species and that of Huegel (though more accurate) is 
scarcely sufficient, I am obliged to re-describe it, and in so doing, take 0, H. 
Sargent's No. 274, and A. Gr. Hamilton's No. 717; the latter is from Perth and 
the former from York, W. Australia, which is in the vicinity of the tj^pe locality. 

Compact shrubs forming drooping masses 2-7 feet long, 1-3 feet in diameter; 
union ball-like. Branches hoary tomentose when young, also the leaves, glabrous 
or almost so when old; leaves opposite, alternate, or in clusters of 3 to 6, terete, 
stout, rugose or wrinkled when dry, 1-G inches long (usually 2-4 inches), straight 
or curved, mostly obtuse, occasionally acute. Inflorescence cymose, axillary, or 
springing from the nodes of the leafless branches. CjTues densely woolh'-, hoary- 
tomentose; the common jjeduncle short and stout, bearing 2 or 3 lateral or partial 
cymes, each with 3 flowers, the central flowers sessile; the lateral flowers on short 
wooUy braeteate pedicels. Bracts fleshy, narrow-uavicular to cordate, acute, 
sometimes exceeding ihe short calyx; calyx cupular, denselj' woolly-tomentose; 
the limb contracted and fully half the lengfh of the calyx, closing over the disc 
after the petals fall. Buds robust, hoary-tomentose, angular in the loAver half, 
clavate at the top, 2-2^ cm. long. Petals 4 or 5, free, reddish inside, thick, the 
upper portion spathulate, concave, with a tuft of rufous deciduous hairs inside 
beneath the acute apex, the lower portion uniformlj' narrow-linear; filaments 



14G THE LORANTHACEAE OP AUSTEALIA^ iv., 

rather thick, compressed, ending in a spur-like callosity, 4-5 mm. from the base, 
fringed on either side with a minute woolly tomentum, the apex jagged and some- 
what eiliate; the free upper portion about 4 mm. long. Anthers adnate, linear- 
oblong, 3 mm. long; the cells conspicuous. Style angular or sulcate, rather firm, 
bent a little below the small, scarcely enlarged stigma. Fruit urceolate, densely 
Loary-tomentose, 5-8 mm. in diameter, contracted at the top into a very short tube; 
disc conical without conspicuous angles. Seeds white, elliptical, with a small 
spongy base, 5-angled towards the top. Hypocotyl white, slender, . verrueose, 5-10. 
mm. long; suctoral disc rather large; embryonic cotyledons narrow-obtuse. 

Bange. — This species, like its congener L. Preissii Miq., has been recorded for 
all the States by Bentham, but I have not seen a typical specimen of L. linop'hyllus 
Fenzl, from South Australia or the Northern Territory. In the compilation of its 
range I have, in some cases, quoted, without prejudice, the records of various 
authors, and it remains for botanists in those States to verify these records. I am 
influenced in my opinion that it does exist in those States, by the range of L. 
Preissii Miq., for it would be unnatural for it to appear in the western and 
eastern States and not in the central States. 

In some of the Victorian and New South Wales specimens the flowers are 
densely-woolly white and almost sessile and, like all other species, the braeteate 
pedicels vary considerably in length. 

Western Australia: Swan River (Preiss. Nos, 1611, 1613, 1615, 1618, quoted 
by Bentham), Perth (on Casuarina glauca), York, Banks of Avon River (on 
Casuarina sp.), Irwin District, near Northampton (on Casuarina sp.), Siberia 
Soak, between Yilgannia claypans and Pindinnie Soak (This may be L. Preissii 
Miq., but the author (Moore) does not state whether his plant is glabrous), 
Slurchison, Shark Bay, Dirk Hartog's Isle, Ernest Giles Range, Sturt's Creek. 
Northern Territory: Hermansburg, Finke River (Trans. Roy. Soc. S.A., v., (1881- 
82), 31, also Report Horn Exped., Part iii., 160), Bay of Rest (quoted by Ben- 
tham, I.e.). Queensland: Evelyn, near Herberton (on Casuarina torulosa Ait.), 
Barakula, via Chinchilla (on "Forest Oak," Casuarina sp.), Warwick, Mount 
Perry, Moreton Bay. New South Wales: Clarence River, (on Casuarina lepido- 
phloia), near Narrabri (on Casuarina Luehmanni), between Boggabri and Narra- 
bri (on Loranthus miraculosus Miq., var. (h), and on Casuarina lepidophloia) , 
New England, Gungal (on C. Luehmanni), Telligerry, Booral, Eugowra (on G. 
Lu,ehmanni) , Narromine (on C. Luehmanni), Wittagoona (on C. lepidophloia), 
Byrock District (on C. lepidophloia), Trig. Station, Louth (on Casuarina sp.), 
Broken HiU, Banks of Campbell's Creek (on Casuarina lepidophloia), Barrier 
Ranges, South-Western N.S.W., Temora-Barmedman (on C. glauca), Lake Albert 
(on C. lepidophloia), Moama (on C. Luehmanni). Victoria: Werribee, Port 
PhiUip, Geelong Distiict, North Creswiek (on "Blackwood-trees," Acacia melan- 
oxijlon, but the mistletoe was not observed on any other plant, nor on tlije smaller 
second growths of the Blackwoods. The elevation is about 1430 ft.), Dumosa (on 
"Bull Oak," Casuarina Luehmanni), Dimboola (on Casuarina glauca), towards 
the Junction of the Ovens River and the Murray (on Casuarina sp.). 

Sargent (Ann. Bot., xxxii., 1918, 216) gives an interesting biological note on 
this &-pecies. 

Affinities. — Its closest affinity is with L. Cambagei, from which it differs in 
its large, wooUy-tomentose buds and flowers, thicker leaves and more urceolate 
and woolly-tomentose fruits. 

In general appearance and in minor characters it resembles L. gihherulm 



BY W. F. BLAKELY. 147 

Tate, from which it is readily distinguished by its triads of flowers, free filaments, 
and less tomeutose leaves. 

Hosts. — Casuarineae: Casuarina glauca Sieb., C. lepidophloia F.v.M., C. Luelv- 
manni R. T. Baker, C. suherosa Ott. et Diet., C. torulosa Ait. Loranthaceae : 
Loranthiis miraculosm Miq. var. Boormani. Leguminosae: Acacia melanoxylon 
R.Br. Myoporaceae: EremopJiila MitckelU Benth. 

17. LORANTHUS cONSPicuus Bailey. (Plate sii.) 

Qland Agrie. Journ., xxvi., 1911, 198, Plate sx.; Comprehensive Cat. Qland 
PL (Fig. 448 bis.), 464. 

A somewhat stout plant; the branches dichotomous, closely and irregularly 
ribbed, often of a glaucous colour between these reddish ribs; the nodes rather 
close and prominent. Leaves coriaceous, nearly or quite sessile; the larger ones 
2^ to 3 in. long, 1^ to If in. broad, obliquely ovate, very blunt at the apex, taper- 
ing to the base from about the centre; margins entire, somewhat undulate, 3 to 
5-nerved, the nerves slightly branched, the reticulation faint; the smaller leaves 
more numerous, oblong, or obtuse cuneate, sessile or very shortly p etiolate, from 
1 in. to 2J in. long; the margins even, usually 3-nerved. Flowers in axillary cymes, 
very dense and dark-coloured, sessile; the bracts and calyx clothed with hairs, 
hoary-white on the calyx, dark on the bracts. Bracts triangnalar; calyx-lobes 
minute. Petals free, 6 to 9 lines long, narrow. 

2Ta6.— Eidsvold (Dr. Thos. L. Bancroft). 

Supplementary notes from specirn^ens received from Dr. Bancroft from the same 

locality. 

Inflorescence axillary, composed of 5 to 9 almost sessile cymes in the axils, 
or springing from the nodes of the old branches, often encircling the nodes. 
Cymes two-branched, each branch bearing three closely sessile flowers. The com- 
mon peduncle hoary pubescent, about 3 mm. long, becoming more elongated in 
fiTiit, in some specimens nearly 1 cm. long and almost glabrous. 

Bracts somewhat spreading, narrow lanceolate, concave, ciliate, nearly glab- 
rous on the back; the persistent bract thicker than the deciduous ones. Buds 
glabrous, narrow cylindrical clavate, acute or obtuse, 15 mm. long. Calyx 
cylindrical, the base densely hoary-tomentose, the rather large limb glabrous or 
nearly so, with an almost entire ciliate margin. Petals 5, greenish, divided to the 
base into narrow-linear spathulate segments, minutely downy along the inner 
margins, above the attachment of the filaments, deflexed after anthesis. Fila- 
ments compressed, gTeen, terminating a purplish erect adnate anther about 1 mm, 
long; style green; angular; stigma someAvhat conical or scarcely capitate. Fruit 
urceolate to bottle-shaped, with a rather long neck, the inflated portion densely 
tomentose, 5 mm, long, 3 mm. in diameter. Epicarp hard and thick; seeds 
globose; viscin scanty. Hypocotyl green, rugose; endosperm green. Embryo 
clavate, the embryonic cotyledons spathulate when opened out, not withdrawn on 
germination. 

Range. — L. conspicuus is a Queensland species, known only from Eidsvold 
and Broad Sound. The author points out that the Broad Sound specimen was 
"received several years ago from the Herbarium, Royal Botanic Gardens, Kew, 
and labelled L. coronatus; no authority, however was given for the name coron- 
atus, which has since been applied by Engler to an African species." 



148 THE LORAXTHACEAE OF AUSTRALIA, iv., 

Affinities. — L. conspicuus is allied to L. Quandang Lindl., as suggested by 
the author, from which it is readily separated by the differently shaped leaves, 
almost sessile cymes in dense clusters, sessile flowers, and very dissimilarly shaped 
buds and fruits. 

Some forms of L. congener superficially resemble L. conspicuus in the leaves, 
but the floral and earpological characters are veiy different. 

L. Betchei has the characteristic inflorescence of this species, but it is con- 
siderably longer; the leaves are also much narrower. 

Hosts. — Loranthaceae : Loranthus Quandang Lindl. var. Bancrofti Bail. 
Rhamnaeeae : AlpMtonia excelsa Reiss. 

18. LORANTHUS Betchet, n.sp. (Plate"xiii.) 

Frutex glaber erectus confertus, ramis- teretibus lenticularibus paullum apud 
nodos trementibus apices juveniles glabros praebintibus. Folia opposita angusta 
ad late-spathulata vel obovata obtusa tripli-costata, petiolata. Inflorescentia 
cymosa, ejTms biradiatis, utroque ramo 3 flores sessiles ferente; pedunculus com- 
munis teres 8-12 mm. longiis parciter cinereo-tomentosus sicut pedicelli bracteati. 
Bracteae acutae navicvdares obscure brunneae. Gemmae glabrae tenues 1 mm. 
diam. parvim elavatae 20-25 mm. longae. Calyx pyriformis 3 mm. longis basi 
pubescens. Corolla 5-petalis liberis viridibus. Antherae adnatae angusto-liueares 
1 mm. longae. Fructus urceolatus cinereo-tomentosus 6 mm. longus 4 mm. diam. 

Glabrous, erect, compact shrubs often of much the habit and general appear- 
ance of L. congener Sieb. Branches terete, lentieulate, slightly swollen at the 
nodes, the young tips quite glabrous. Leaves opposite, narrow to broad spathu- 
late or sometimes elliptical in the Queensland foiins, obtuse, triplinerved, the 
veins more prominent underneath and slightly depressed on the upper surface, the 
lateral ones obscure or wanting, usually drjdng a dark colour like L. congener, 
4-8 cm. long, 1-3 cm. broad, tapering into a well defined petiole. Inflorescence 
eymose, cymes two-branched, each branch bearing 3 sessile flowers. The common 
peduncle terete, 8-12 cm. long, slightly hoary-tomentose, as well as the 4 mm. 
bracteate j^edicel. Bracts acute, navicular, dark brown, somewhat cornose, almost 
glabrous, the central one persistent, the lateral ones deciduous, similar in every 
respect to those of L. conspicuus, but more glabrous. 

Buds glabrous, slender, about 1 mm. in diameter, slightly clavate, the de- 
marcation lines of the segments shghtly raised, giving the buds an angular ap- 
pearance when dry, 20 to 25 mm. long. Calyx pear-shaped, 3 mm. long, pubes- 
cent at the base with a hoary and rufous tomentum, the thin entire limb almost 
glabrous except for the minute red-brown ciliate border. Petals green, usually 5, 
free, marked along the entire Icng-th of the inner margin with a minute pubes- 
cence. 

Filaments straw-coloured, compressed, attached to the centre of the segment; 
anthers adnate, narrow-linear, about 1 nmi. long. Style angular, with a scarcely 
enlarged stigma; disc raised around tlie base of the style. Fruit urceolate, hoary- 
tomentose; except the somewhat long neck, 6 mm. long, 4 mm. in diameter. Seeds 
globose, similar to those of L. conspicuus; the cotyledons are probably tlie same 
a.s in that species. 

Ballina, parasitic on Alpliitonia excelsa, W. Bauerlen, No. 042, 12, 1892. The 
type. 



CY W. F. BliAKEIA'. 



149 



Like L. conspicuus Bail, this species is found on the brush timber of the 
North Coast, extending from Ballina to North Queensland. 

Jlange. — New South Wales: Ballina. Queensland: Cairns, Yarrabah, near 
Cairns, Howick Group (refen-ed to by Bentham in B.Fl. under L. odontocalyx 
F.v.M.), Buudaberg. 

Affinities. — L. Betchei has the characteristic inflorescence of L. conspicuus, 
but the common peduncle is much longer, exceeding 10 mm., whereas in L. con- 
spicuus it is very short, almost obscure. The buds too, are longer, being twice 
the length of those of L. conspicuus, while the leaves are usually narrow oblong- 
cuneate, and fairly constantly triplinerved. Those of L. conspicuus are often 
quinquenerved, and on the whole coarser and more venulose. 

The differences in the characters appear to me to be sufficiently distinct as to 
warrant specific rank. The Ballina and Cairns specimens, in spite of the great 
distance between the tAvo localities, are inseparable, and lead me to believe that 
the characters are stable. 

L. Betchei is similar to L. congener Sieb. in appearance; the two-branched 
c3Tiie, smaller buds, sessile flowers, and different-shaped bracts are the points by 
"v\ hieh it can be distinguished from L. congener. 

Named in honour of the late Ernst Betche, late Chief Botanical Assistant, 
Botanic Gardens, Sydney. 

iTos^s.— Rhamnaceae : Alphitonia excelsa Reiss. 

(a) Var. dubia_, n.var. (Plate xiii., fig. 7.) 

Rami graciles teretes juveniles aliquanto minuta farinaceo ferrugineo tomen- 
to induti. Folia opposita vel alternata ovato-laneeolata in brevem petiolum re- 
pente constricta obscure tripli-costata vel latiora quinque-costata 3-4 cm. longa, 
1.5-2 cm. lata exsieeata obscure jjlumbea. Infloreseentia est cjtuus singularis pen- 
dulus duobuo ramis ex arilla ortus. Flores non visi. Fructus juvenilis sessilis 
intra latas fere cordatas _ bracteas, dense cinereo-tomentosus atque magis cylin- 
dricus quam in L. Betchei. 

Branches slender, terete, the young ones more or less covered with a minute, 
mealy ferruginous tomentum. Leaves opposite or alternate, ovate-lanceolate, ob- 
long to elliptic, abruptly contracted into a short terete petiole, obscurely trip- 
linerved or occasionally a few of the broader ones quinquenerved, 3 to 4 cm. 
long, 1.5 to 2 em. broad, drying a dark lead colour; the petioles 3 to 7 mm. long. 

Inflorescence eymose, consisting of a solitary 2-branched erect or pendulous 
axillary cyme. FloAvers not seen. The specimen is in juvenile fruit only, which 
are sessile within the broad almost cordate bracts, and densely hoary-tomentose, 
and more cylindrie in outline than those of L. Betchei. 

Welsh River, T. Barclay Millar, March, 1891, Queensland Hei-barium, kindly 
lent by Mr. C. T. White. 

Perhaps a distinct species, resembling L. Betchei in the two-branched cj-me 
and in the sessile ai*rangement of the inflorescence. The bracts are also broader 
than those of L. Betchei-, and the tomentum on the young branches is another 
character which differentiates it from that species. 

No doubt when more material of this plant is available, it will be found to 
possess other characters which will distinguish it from L. Betchei more definitely. 

(&) Var. TOMENTiLLA, n.var. (Plate xiii., figs. 8-10.) 

Rami aliquanto cxassi partibus juvenilibus minute cinereis. Folia petiolata 
late-elliptica coriacea rugosa 3-5 costata ut plurimum 4-6 cm. longa 3 cm. lata. 
Cymi 3-8 in fascieulis bi ramati ut in forma typiea. Gemmae per contra baud 



150 THE LORANTHACEAE OF AUSTRALIA,, IV., 

glabrae sed minute einereo-tomentosae. Bracteae quoque latiores et magis car- 
nosae. 

Branches rather stout, the young- parts minutely hoary; leaves broadly el- 
liptical, coriaceous, rugose, 3 — 5-nerved, the largest 4-6 cm. long, 3 cm, broad, 
petiolate. 

Cymes in clusters of 3-8, two-branched as in the typical foi-m, but the buds 
minutely hoary-tomentose, instead of glabrous; the bracts are also broader and 
more earnose. 

Sweers Island, Gulf of Carpentaria, J. F. Bailey, 6.1901. In Queensland 
Herbarium. I have only seen one specimen, but it seems so distinct from L. 
Betchei that I have no hesitation in giving it a name. 

It is a coarser and apparently larger plant than var. dubia, and also differs 
from it in the clustered cymes. 

19. LOKANTHUS OBLiQUA^ n.sp. (Plate xiv.) 

Frutex glaber plerumque ramis teretibus pallidis. Folia opposita sessilia ob- 
longo-euneata ad elliptica aliquanto obliqua 3-costata 2-4 cm. x 1.5-2 cm. In- 
floreseentia cymosa cymis biradiatis utroque radia 3 flores centralem sessilem 
praebente. Gemmae graciles teretes acutae clavatae 20-23 mm. Calyx eylindrica. 
Corolla 5 petalis, liberis. Filamenta eompressa. Antherae lineares adnatae. 
Stylis tenuis angnilaris. Stigma sub capitatum. Fructus urceolatus. 

Glabrous except for a minute rufous tomentum in the axils of the leaves and 
at the base of the inflorescence. Branches terete, slightly swollen at the nodes, 
pale coloured. Leaves opposite, sessile, but not stem-clasping, oblong cuneate to 
elliptical, somewhat oblique, coriaceous, 3-nerved, much paler on the lower sui'- 
f ace, 2-4 cm. long, 1.5 to 2 cm. broad. Inflorescence cjnaiose, cymes two-branched, 
each branch with 3 flowers, the central one sessile, the lateral on pedicels 4-5 mm. 
long ; common peduncle slender, terete, gTadually thicker towards the top ; brac- 
teate peduncles ciliate. Bracts narroAV, navicular, the minute membranous mar- 
gin ciliate with rufous hairs. Buds slender, terete, acute, clavate, 20-23 mm. long. 
Calyx cyKndric, glabrous, except for the entire ciliate limb, 2-3 mm. long. Petals 
free, usually 5, acute; filaments compressed, attached about the centre of the! 
petals; anthers linear, adnata. Style slender, angular, usually bent in the upper 
portion; stigma sub-capitate. Fruit ureeolate. 

Johnstone River, North Queensland, Dr. T. L. Bancroft. Labelled L. Quan- 
dang Lindl. var. amplexifolvus in Brisbane Herbarium. The type. 

It has also been collected at Yarrabah, near Cairns, parasitic on Mangroves. 
This specimen is in fruit, which is very similar to the fruits of L. congener Sieb. ; 
the leaves are also broader, more ovate and less oblique. Both localities are on 
the North Queensland coast. 

Affinity. — In the inflorescence L. obliqua shows a strong affinity to L. Benth- 
ami, but the leaves of both species are very dissimilar. It has much in common 
with L. Betchei, from which it is distinguished by its almost sessile leaves. 

Host. — Verbenaceae: Avicennia officinalis L. 



BY W. F. BLAKELY. 151 

EXPLANATION OF PLATES III.-XIV. 
Plate iii. 
Loranthiis queenslandicus, n.sp. 
1. Portion of flowering branch (nat. size); 2. Bud, enlarged; 3. Flower, en- 
larged; 4, 5. Bracts, showing variation; 6. Calyx, showing the depressed disc; 7. 
Fruit (nat. size) . 

Plate iv. 

A. Loranthus Cycneus-Sinus, n.sp. 

1. Flowering branch (nat. size); 2. A bud; 3. Flower; 4, 5. x\nthers, front 
and back view; 6. A spent calyx with the rather thick style. 

B. Loranthits Mackayensis, n.sp. 

7. Flowering branch (nat. size); 8. A bud; 9, 10. Anthers, front and back 
view; 11. Calyx with style attached. 

Plate V. 
Loranthus miraculosus Miq. 
1. Flowering branch (nat. size); 2. Bud, enlarged; 3. Flower, enlarged; 4. 
Anthers, enlarged; 5. Fruit, slightly enlarged; 6. Lower portion of segment. 

Plate vi. 
Lorantliits miraculosus Miq., (a), var. Melaleucae (Tate), uov. comb. 
1. Flowering branch (nat. size); 2. Bud; 3. Lower portion of segment. 

Plate vii. 
Loranthiis nviraculosus Miq. (b). var. Boormani, n.vai". 
1. Flowering branch, Cobar (L. Abraham, No. 774); 2. A bud; 3. Fruit (nat. 
size); 4. A germinating seed with two radicles; 5. Embryo, enlarged; 6. Embryo 
with the cotyledons opened out; 7. Lower portion of segment showing the basal 
spur, which is more prominent on some segments than others; 8, 9. Bud, cah^x 
and style of (c.) vds.piibigera, n.var. 

Plate viii. 
Loranthus Gaudichaudi D.C. 
1. A young flowering plant (nat. size); 2. Bud, enlarged; 3. Flower, en- 
larged; 3a. Base of segment showing callosity, enlarged; 4. Anthers, front and 
back view; 5. Bracteate pedicel; 6. Fruit (nat. size, and shghtly enlarged); 7. 
Seed (nat. size); 8. Seed, enlarged; 9. Germinating seed with two radicles; 10. 
Germinating seed, enlarged; 11. An advanced seedling showing the linear-obtuse 
primary leaves. 

Plate ix. 

Loranthus Preissii Miq. 

1. Flowering branch (nat. size); 2. Bud, enlarged; 3. Flower, enlarged; 4. 

Segment, enlarged; 5. Fruit, enlarged; 6. Seed (nat. size); 7. A young seedling 

with the endosperm removed showing the cotyledons; 8. x^n advanced germinating 

seed, enlarged; 9. var. d-idyma, n.var. 

Plate X. 
Loranthus Camhagei, n.sp. 
1. Flowering branch (nat. size); 2. Bud, enlarged; 3. Flower, enlarged; 4. 
Base of segment, enlarged; 5. Fruit (nat. size); 6. Germinating seed, enlarged; 
7. A seedling (nat. size), probably in its second year. 

Plate xi. 
Loranthus Imophijlhis Fenzl. 
1. Flowering branch (nat. size); 2. A triad of sessile buds; 3. A bud, en- 
larged; 4. Flower, enlarged; 5. Bracts, enlarged; 6. Lower portion of segment 
showing the spur-like callosity; 7. Segment of a flower from Queensland, showing 
a slightly hirsute filament and different-shaped callosity at the base; 8. Fruit (nat. 
size) . 



152 THE LORAXTIIACEAE OF AUSTEALIA, iv. 

Plate xii. 
Loranthus conspicuvs Bail. 
1. Flowering branch (nat. size); 2. Bud with deciduous bract attached; 3. 
A cymule showing 3 buds, and 3 calyces; 4. F'lower; 5. Anthers, front and back 
view; 6. Fruit (nat. size); 7. Fruit, enlarged; 8. Germinating seed, enlarged. 

Plate xiii. 
Loranthus Betchei, n.sp. 
A. L. Betchei, n.sp. 
1. Flowering branch (nat. size) ; 2. Bud, enlarged; 3. A segment with stamen at- 
tached; 4. Anther; 5. Style; 6. Two leaves and fruit, Howicks Group (F. 
Mueller). 

B. L. Betchei, var. duhia, n.vai'. 

7. Fruiting twig (nat. size). 

C. L. Betchei, var. tomentilla, n.var. 

8. Leaf and cyme (nat. size); 9. Bud, enlarged; 10. A broad leaf, Sweer's 
Island (J. F. Bailey). 

Plate xiv. 
Loranthus obliqua, n.sp. 
1. Flowering branch (nat. size) ; 2. Common peduncle and secondary peduncles; 
S. Bud; 4. Flower; 5. Anther; 6. Fruit. 



SOME NOTES ON THE PERMO-CARBONIFEROUS AND OVERLYING 
SYSTEMS IN CENTRAL QUEENSLAND.* 

By H. I. Jensen^ D.Sc. 

[R:ad 18lh April, 1923.] 

1. Introduction. The writer has for the past eighteen moaths been carry- 
ing on geological reconnaissance work in the country lying between the Charle- 
viUe Railway line and the Longreach Railway line in Western Queensland. He 
is in consequence in a position to offer some remarks on the sequence of forma- 
tions within the area. An advance account of the general geology of the region 
has already been published by the writer in the Queensland Government Mining 
Journal for the months of March, August and October, 1921. The present paper 
is only a brief summary of results. 

2. The Geological Sequence in the Carnarvons. The entire area examined 
by the writer is for the sake of convenience referred to as the Carnarvons. The 
rocks represented in the area include volcanic, plutonic and sedimentary rocks 
ranging from metamorphic to recent. The igneous rocks it is not proposed to 
discuss here, except that it might be stated that the volcanic rocks of the Car- 
narvon Range, Dividing Range and Springsure areas are chiefly Tertiary, the 
only other volcanics noted being those of Lower Bowen and Star age, together 
with some tuffy sedimentaries in the Ipswich Series. 

The oldest sedimentary rocks are metamorphic schists and serpentines of 
undetermined (perhaps Devonian) age, outcropping over a few small isolated 
areas as buried pinnacles recently exposed by denudation. 

The next rocks in point of age are those of the Star Formation, which are 
taken to be Lower Carboniferous ranging to Middle Carboniferous. The Lower 
Star rocks are sandstones with abundant Lepidodendron veltheimianum and inter- 
bedded basic lavas. 

At Mt. Budge, Drummond Range, a specimen of Aneiniites austrina has been 
obtained and the higher Star Beds are mudstones and shales of marine origin, 
with abundant fish remains. The Star Beds are considerably folded, and are 
the equivalents of the Burindi Series in New South Wales. 



*This paper is published with the understanding that the views expressed 
are those of the author and are not necessarily endorsed by the Queensland 
Geological Survey by whom he was employed at the time of making the observa- 
tions. — Ed. 



154 PERMO-CAEBONIFEROUS AND OVERLYING SYSTEMS IN CENTRAL QUEENSLAND, 

Lying uneouformably on the Star Beds we ba.ve the Lower Bowen Series, 
outcropping along the Nogoa River from Nandowrie Peak to Springsure. It 
consists of conglomerates, largely glacial, tillites, shales (including some highly 
contorted varve-shales), volcanic rocks (both basic and acid, with abundant preh- 
nite and other zeolites) and, forming the top of the formation, a limestone con- 
taining an abundant fauna, including Euryclesma cordata. These rocks are the 
equivalents of the Kuttung Series. 

Conformably resting on the Lower Bowen Series we have the Middle Bowen, 
consisting of a lower conglomeratic sandstone, a middle formation of shales with 
important coal measures (Consuelo Measures), and an upper sandstone with fre- 
quent red ferruginous sandstone bands. The coal beds and shales contain both 
Gl-ossopteris and Gangamopteris cyclopteroides. The Middle Bowen Series 
fringes the Lower Bowen on the south and was traced with an approximate east- 
west strike and general southerly dip at gentle angles from Blue Hills to near 
Springsure. 

South of the Middle Bowen Series and lying conformably on them we have 
the Upper Bowen consisting of basal limestones, overlain by coal measures with 
fine coal and kerosene shale seams, these again overlain by shales containing a 
rich flora of Glossopteris and Gangamopteris angvstifolia, and then a sandstone 
formation (the Clematis Sandstone) which passes upwards into the basal mica- 
ceous sandstones of the Ips\¥ieh Formation. The top stages of the Upper 
Bowen, as well as the basal strata of tht Ipswich, contain abundant stems of 
ScMzoneura, Neocalamites, Phyllotheca and other Equisetales, which seem to be 
also very abundant in the Tolmies-Clermont Coal Series, and are rather charac- 
teristic of the passage beds between the Permian and Triassic. 

The Ipswich Series consists of the basal micaceous sandstones, over which 
we find a thickness of freshwater limestones with abiindant ScMaoneura, and 
then tuffs, shales and sandstones intercalated. The tuffs are often olive-green 
and the shales in part chocolate-coloured. They contain abundant remains of 
Equisetales and of Thinnfeldia odontopteroides. In the writer's opinion, this 
series is the equivalent of the Narrabeen Series in New South Wales 

Overlying the Ipswich Formation conformably, but commencing with a 
coarse conglomerate, we have the Bundamba Formation which is lithologically 
similar to the Hawkesbury Sandstone and is, in the writer's opinion, the equiva- 
lent in age of that well-known Sydney formation. 

Above the Bundamba, and conformably resting on it, we have the Walloon 
Series, divisible to the north of Roma into a basal, felspathic sandstone, a highly 
calcareous freshwater formation, consisting of limestones, shales, calcareous Sand- 
stones, coals and kerosene shales, then a thick sandstone, then more calcareous 
sandstone, sandstone, felspathic shales and carbonaceous beds closely inter- 
laminated, forming the Orallo Coal Measures, then more shales and sandstones 
with minor inferior coal-seams to the top. 

The Walloon Fonnations north of Roma have yielded no Thinnfeldia, and 
no large Taeniopterideae. Cladophlehis australis, Taeniopteris spathulata, Oto- 
zamites Feistmanteli, and Sphenopteris superha were the principal fossils found 
in the Walloon in this area, the only others being fragments of Neocalamites and 
other Equisetales, Plilophyllum pecten, Dictyophyllum Davidi, and seeds and 
spores. 

The Walloon Formations are Jurassic and are tlie equivalents of the Upper 
Clarence, and of the formations met with by the wiiter west of the Nandewar 
and Wamimbun^le Mts. in New South Wales. Gilgai country like that in the 



BY H. I. JEKSEN. 155 

Narrabri corner of the Pilliga scrub, N.S.W., is one of the features of Lower 
Walloon physiography. Mound springs occur in the upper part of the Bun- 
damba at Mt. Hutton, Timor and Crystal Brook and are comparable with the 
"Mound springs in the Pilliga Scrub, N.S.W. The plain country, commencing 
west of the Nandewar Mountains, is of the Queensland Walleon type and the 
Walloon Formations probably swing west through the northern edge of the 
Pilliga Scrub, then south-west through the western rim of the Pilliga Scrub, 
and thence south to Tundebrine and Tenandra. 

Overlying the Walloon conformably we have the Lower Cretaceous (Cretace- 
ous marine), the basal beds of which are exceedingly rich in fossil wood, often 
Teredo-bored, mixed with marine shells. It is at the top of the marine series 
we get the first marked unconformity since the inception of the Lower Bowen. 

On the Marine Cretaceous Ave have the Cretaceo-Tertiary (Desert Sand- 
stone) lying unconformably and horizontally stratified. 

The Cretaceous Marine Series is gently folded after the same manner as the 
Walloon, as described in the Queensland Government Mining Journal, August, 
1920, into gentle anticlines on the direction of general dip, and terraces on the 
direction of general strike. 

3. The Bowen Formations in the Type District. On a subsequent journey 
from Mt. Coolon to Bowen, the writer noticed a strong development of dacites, 
felsites, rhyolites and porphyries, apparently of Upper Devonian age, at the 
head of the Suttor River, a tributary of the Burdekin. These volcanics are 
lithologically the equivalents of the Snowy River Porphyries and the Macedon 
dacites. In age they are probably Devonian. 

The Bowen Formations in the type district lie in a synclinal basin having 
its centre near Havilah, the Lower Bowen Formations outcropping near the edge 
of the basin. The coal now being developed at the Bowen Coalfield is a 
Glossopteris-heaxmg freshwater series, overlying a series of boulder beds and 
conglomerates interbedded with sandstones and volcanics, and underlying the 
lower of the two marine formations described by Dr. Jack in his Middle Bowen 
Series.* The marine limestones overlying the Bowen Coalfield (CoUinsville coal 
measures) are lithologically and in general fossil assemblage very like the Dilly 
limestone (Lower Bowen) north of Springsure. The marine limestone at Pelican 
Creek (Bowen Coalfield) is immediately underlain by a glacial bed with faceted 
boulders and right in the marine limestones we have huge granite erratics carried 
by floating ice and dropped in the midst of delicate fossil organisms. Dr. Jack 
in his original description of the field noticed these strong evidences of glacial 
action. The evidence of glaeiation below, and within the marine limestones both 
at Dilly and Pelican Ck., confirms the belief that we are dealing with the same 
geological horizon and that Eurydesma will be found in the Pelican Ck. lime- 
stone. This is also supported by the Carboniferous affinities of the fossils in 
each of the two localities. Thus the Pelican Ck. Limestone contains, inter alia, 
abundant Productiis cora, Productus sub quadratics, Stenopora australis, and Pro- 
toretepora ampla var. woodsii. Of these, all except Productus cora have also 
been recorded from Dilly near Springsure, and Productus cora has been recorded 
by Mr. B. Dunstan, inter alia, from limestone lying west of Chine Bluff, Dawson 



* Preliminary Report on the Bowen Coal Field by R. L. Jack, 1879. See 
also Geology of Queensland by Jack and Etheridge. 



156 PER3I0-CARB0NIFER0US AISTD OVERLYI>rG SYSTKilS IN CENTRAL QUEENSLAND, 

^'alley, a limestone formation wliich Mr. Dunstan found continuous with "Grym- 
pie" limestone (see p. 8, Geology of the Dawson and Mackenzie Rivers, etc., 
Qland. Geol. Surv., Pub. 155, 1901). These fossils serve to bring together the 
Lower Marine of Dilly, the Lower Marine of the Dawson-Mackenzie area and the 
Lower Marine of the tyi^e district and show that this marine series is Upper 
Kuttung or Lower Uralian and that the Bowen coal is a Carboniferous coal, 
lower than the Greta coal of New South Wales. The Bowen (Collinsville) coal 
is practically of au interglacial age. 

In the type district the coal seams corresponding in position to those of the 
Greta of New South Wales, the Consuelo coal of the Carnarvons, the Bluff and 
Dawson coals in the Dawson-Mackenzie area, are apparently the "Daintree" seam 
and also the associated series which lie above the Lower Marine. In the fresh- 
water series between the Lower and Upper Marine horizons on the type field, 
Glossopteris and Gangamopteris cyclopteroides with Sphenopteris, Cladophlebis 
Eoylei and Pliyllotheca were very abundant. 

4. Glaciation and Folding. The period of maximum glaciation in Queens- 
land, when the numerous thick boulder beds of the Lower Bowen were deposited, 
was the period of great movements of folding, and was a period immediately 
fdllowing an age of great granitic intrusions. 

On palaeontoiogical • e^ddence we must regard the glacial period as Upper 
Carboniferous, and the last period of granitic intrusions in Palaeozoic times as 
Middle or Lower Carboniferous. This holds for the Carnarvons, Clermont, Mt. 
Coolon, the Drummond Range and other parts of Central Queensland and is 
equally true for the Maekay-Bowen Districts and North Queensland generally. 
Thus in the Chillagoe-Herberton area the newer granites are later than the 
Devonian Hodgkinson Series, but earlier than the undisturbed Permo-Carboni- 
ferous Mulligan Series. 

The principal period of intense folding and mineralisation in Queensland 
was the Lower to Middle Carboniferous, though the processes appear to have been 
a little later in South Queensland than in the north of the State. 

5. Absence of Glacial Evidence in the Middle and Upper Boiven Series. On 
tlie writer's many traverses across the Bowen strata in the Carnarvons he failed 
to see anywhere any sign of a glacial fonnation equivalent to the Branxton. 

There was no evidence of any glaciation later than the Lower Bowen. Con- 
siderable importance must therefore be attached to the glacial series. They 
should be a more definite landmark in time than the evidence of more or less in- 
complete palaeontoiogical collections. 

Shell-banks vary in the content of living species in different i^laces on any 
modern shore line. So they did in all ages. Until exhaustive palaeontoiogical 
collections have been made, judgment passed on the age of beds on such fossil 
collections is likely to be weakly founded, especially in an area lilie the Bowen 
Coalfield where two nuirine series exist, and the fossils have in many cases be- 
come mixed and fiekl-woi-k has not yet definitel.y delineated which limestone the 
fossil locality belongs to. The Pelican Ck. limestone lias, however, been de- 
finitely determined to be tli(; Lower Marine ol' the field as described by Dr. Jack. 

6. Ganganiopt'(!ris and Glossopteris of tcider range than supposed. The 
Lower BoAven top limestone horizon being on the strength of glacial evidence 
probably the equivalent of tlie Dilly Springsure horizon, and the Dilly fossils 
))eing accompanied by Eurydef<ma we see that the Bowen coals are Upper Car- 



BY H. I. JENSEN. 



157 



m m 



fOf 



r 



^ 



CO 

5^ e 



1w 



<B 



!3^ 



CJ a) 









P^ 2? 



^ o 






^ O 




> SH 


r^ 


'3 




^ 


i* 


m 


a 








5* 



OS 



« ^ 



oi u 

Sp 






O) 



Oi 






m 


■-a 


m 


P 










^ 


^ 


o 


5>D 


p 


!3 

O 




r^ 


ba o 


fl ^ 


•rH 


V 




fH 


O 


CS 


7«- '^ 


'TS 




i=l 




c3 


^ 




a> 


O 

O 


> 
o 



o 









■s 



DQ 



eSCQ 

p^ 



13 


d 




o 


n3 


p 


o 
o 


bfl 


^ 


.s 


OJ 




o 


(33 


p^ 


P 



o 


^ 




O 


-y ?-i 




■? g 


o 

a 


-— --^ cS 


d „, 


O 4) 


•^ 2 








o JG 


jd '^ '« 


-i-J r^ 


,^ .•^ c5 


cc ":«. a 


alloon (J 
Upper W 
Orallo G< 


ig Sand 
alcareoii 
asal Sai 


mop 


> 




t^ 








C/2 o 

'5 r=) '-^ 

P^ 



P 






cq 



&s ^ fe5 (*! 






iC ryj W 

V Qi r\ 
'^ '^ M 

cS cS s3 

O O f-i 

'^^.§ 

las 



a 5 



-^ 
rt 



CD 



« o 

G O 
Ol 



1p§p^ 

•r Ol '^ X 



t^ -a 



= Ph 



5 H 



a) 0! « 

S u ^ 



158 PERMO-CARBONIFEROUS AND OVERLYING SYSTEMS IN CENTRAL QUEENSLAND- 

boniferous, and it follows that Glossopteris commenees in the Carboniferous. It 
is equally interesting to note that Gangamopteris cyclopteroides is very abundant 
in the Middle Bowens and G. angnstifolia in the Upper Bowens. 

7. StratigrapMcal Sequence. The accompanying table has been con- 
structed to show the stratigraphical sequence and thickness of the beds examined 
— the equivalents in other geologically examined parts of Queensland, the New 
South Wales equivalents and the type fossils. 

8. Afterword. It will be noticed that the Bowen Formations have been 
dealt with in this paper in the original sense, this being more in harmony with 
observed facts in the Carnarvon area than the classification of the Bowen For- 
mations adopted by Mr. Dunstan in his Dawson-Mackenzie report. Mr. Dunstan 
varied from Dr. Jack in speaking of a Lower Bowen consisting of Lower Marine, 
overlain by Lower Freshwater (Coal Measures), then Upper Marine and Upper 
Freshwater. This group of formations was bracketed as Lower Bowen and the 
Clermont-Tolmies Coal Measures were classed as Upper Bowen, the Middle 
Bowen of Dr. Jack not being recognised. 

The present writer has varied from Dr. Jack's conception of the Bowens 
only in one particular. Finding that the Upper and Lower Marine limestones 
are only comparatively thin formations, the one lying between the Lower and 
Middle formation, the other between the Middle and Upper Bowen, it was deemed 
advisable to put the Lower Marine into the Lower Bowen Series where it also 
fits naturally on glacial evidence, and to put the Upper Marine into the Upper 
Bowen (Permian) where it fits well on account of the abundance of Strophalosia. 



159 



REVISION OF THE GENERA ETRON, CISSEIS AND THEIR ALLIES. 

{BUPRESTIDAE.) 

By H. J. Carter, B.A., F.E.S. 

(Twelve Text-figures.) 

[Read 30th May, 1923.] 

Th€ genera Ethon and Cisseis were first published by Castelnau and Gory in 
their famous "Monograph." Their diagnoses of these genera, together with the 
list of species, were so confused as to require considerable modification at the 
hands of later authors; e.g., these genera were distinguished, inter alia, by the 
alleged differences in the respective number of joints of the palpi; while of the 
eight species placed by them under Ethon, four, at least, are now' referred to 
Cisseis. For a complete diagnosis of Ethon and of Cisseis see Kerremans (Genera 
Inseetorum, 1902, pages 226 and 227). 

Ethon is clearly differentiated from Cisseis by two characters in combina- 
tion: (1) a strongly bilobed head, (2) the clearly longitudinal punctate elytra. 
The claws of their tarsi are long and bifid, somewhat as in Neospades, appearing 
as four equally long sharp spines. The posterior basal joints are, however, more 
elongate than in Neospades, the basal joint being the longest, while joints 2, 3 
and 4 are of nearly equal length The nomenclature has been complicated by mis- 
identification and synonymy and the following table and notes on the species will, 
it is hoped, simplify the study by our future workers. 

Both Ethon and Cisseis occur on flowers of many species. I have taken E. 
corpulentum and E. affinis chiefly on Dillwynia, Pultenea and Leptospermix/m in 
the Sydney region. 

Froggatt (Australian Insects, p. 165) states that "E. affinis forms galls upon 
the stems of Pultenea stipularis," while E. corpulentum and E. marmoreum (? 
Cisseis acuducta Kirby) "make rounded gaUs upon the roots of Dillwynia erici- 
folia, sometimes as many as 20 on one plant, clustering round the base of the 
stem." 

No. of names No. of species No. of new 

investigated considered valid species added 

1 Ethon 12 o 1 

2 Cisseis 86 38 7 

3 Neospades 17 10 — 

4 Hypocisseis 13 5 2 

5 Alcinous 2 1 — 

Total 130 59 10 



160 REVISION OF THE GENERA ETHON, CISSEIS AND THEIR ALLIES, 

Thus after rejecting names that are either preoccupied, synonyms, or nomina 
iiuda, 130 names are reduced to 59, while 10 new species are here recorded. 
Future investigation may suggest some modification of this treatment; but at least 
clearer identification of existing species and their correct nomenclature is now 
possible. 

Table of species of Etlion. 

1 6 Colour bronze or coppery with pubescent fascicles. 

2 4 Form oblong oval, underside subglabrous. 

3 Surface nitid, elytral punctures scratch-like ajfine C. & Gr. 

4 Surface duller, elytral punctures coarser roei Saund. 

5 Fonn sub-conical and convex, underside strongly pilose, corpulentum Boh. 

6 Form shortly ovate, underside subglabrous hreve, n.sp. 

7 Colour opaque violet, form elongate-conic, striation of elytra sub- 
continuous maculatum Blkb. 

8 Colour bronze-green without fascicles fissiceps Kirby. 

Synonyrfiy. (a) affine C. & Q-. = aurifluum Hope = purptirascens Hope 
= proxima Boh. = reichei Chev. 

(6) fissiceps Kirby, (?) Boisd. = virid-e C. & G. = diversum Kerr. 

(c) corpulentum Boh. = fissiceps C. & G. {nee Kirby). 

(d) roei Saund. = suhfasciatum Saund. 

I have seen the types of all except proxima, viride and corpulentum. It is. 
quite evident that fissiceps was wrongly determined in the Monograph of Castel- 
nau and Gory, and equally evident that the figure given by them as fissiceps is 
that of corpnlentum Boh., having the distinguishing oblique fasciation at the 
apex. 

(fe). obviously identical. 

(d). I think suhfasciatum is the male of roei — the types being respectively 6 
and 5 of the common species from Western Australia, variable in size, and the 
pubescence readily abraded — as in the type of roei — which is almost uniform 
bronze with a diUler surface than in affine. 

The species corpulentum Boh. and affine C. & G. are common on the flowers 
of Dillwynia, near Sydney, in September and October. The former is easily 
separated (i.) by its strongly pilose u.nderside, (ii.) by apical markings; a sub- 
apical zig-zag fascia, and behind this, two obliquely outwardly directed markings. 
E. affine has a wide distribution through Eastern and South-East Australia. E. 
maculatum Blackb., type in British Museum, measures 9J x 4 ram., though given 
as 3 3/5-4 2/5 Unes. This apparently is found in New South Wales and Queens- 
land as well as in South Australia, from examples I find in collections, and 
is recognised by the combination of conical form, dingy, vaguely-spotted, violace- 
ous upper surface — the pubescent fascicles being distributed more or less over 
the whole elytra — with nitid, subglabrous underside. E. scabiosum Boisd., mis- 
spelt scahiorum in Castelnau and Gory's Monograph, is quite unknown to me — 
though probably a Cisseis, if the figure of the "Monograph" be correct. 

The following species is undescribed: 

Ethon breve, n.sp. (Text-fig. 1.) 

Shortly ovate, convex. Head, prothorax and abdomen coppery, elytra purple- 
violet with three undefined fascicles on apical half and sub-obsolete fascicles on 
basal half; underside nitid and feebly tomentose, sides of abdomen albo-squamose. 

Head strongly bilobed. 



BY H. J. CARTER. 161 

ProtJiorax short, strongly widened from apex to base; auteiior 
angles obtusely rounded, posterior acute, base strongly sinuate, the 
medial lobe widely oval, surface more finely sculptured than in E. 
roei Saund. or maculatum Blkb. with little transverse sculpture. 
Scutellum large, triangular and smooth. Elytra widest near base, 
thence arcuately narrowed to apex, seriate-punctate, the punctures 
linear, less elongate than in maculatum Blkb., narrower than in roei 
Text-fig. 1 Saund. Underside finely, closely punctate. Dim. 6-7 x 3-3.3 mm. 

Hah. — Western Australia: Perth (J. Clark, and Brit. Mus.), 
Wyndham (J. Clark). 

A species I cannot reconcile with described species — through its shorter oval 
form, nitid underside, the short prothorax — strongly narrowed in front. Types 
in Coll. Carter. The sexes taken in cop. by Mr. Clark. 

Var. (?). — Three examples taken by Mr. A. H. Elston at Mt. Lofty, near 
Adelaide (S.A.), have the head and pronotum dull coppery bronze, the elytra 
purple-bronze, and are more elongate-ovate (narrower) than the Western forms. 
6-7 X 2:^-3 (vix) mm. These have the fine sculpture of hreve, but may prove to 
be a distinct species. 

C I s s E I s Castelnau et Gory. 

Six species were originally named in the Monograph of the above authors, 
namely, duodecim-guttata Guer., albosparsa C. & G., stigmata C. & G., irrorata 
Hope, marraorata C. & G., and cupripennis Guer. ; while, as noted above, at least 
four of the species included under Ethon belong to Cisseis. These are leucostic- 
tum Elirby, hicolor C. & G., marmoreum C. & G. and maculatum C. & G. West- 
woodi was described as a Coraebus. This list omitted cruciata F. (for which see 
Neospades), duodecim-maculata F. and acuducta Kirby. Duodecim-guttata Guer. 
(also Boisd.) is the well known 12-maculata F., of which I have examined the 
type in the British Museum. Albosparsa C. & G. is familiar as a widely distri- 
buted Queensland species. 

Stigmata C. & G. is a well known Western Australian species. 

Irrorata Hope was incorrectly determined by Castlenau and Gory, and is the 
species described by these authors as marmorata, and by Saunders as simiUs, 
while irrorata C. & G. is, I consider, the same as maculata C. & G. I have seen 
the types of Hope and Saunders. The species is well known from New South 
Wales and Victoria (see infra). 

Cupripennis Guer. (also Boisd), described from Port Jackson, is a common 
species in South-Eastern Australia, and generally well known in Museums. 

Leucosticta Kirby is widely distributed over the greater part of Australia, 
associated with Acacia foliage. As in all common species, there is great varia- 
tion in size and colour. The type measures 14 x 5 mm., has a brassy pronotum 
and green underside; but green often replaces the brass of the prothorax and 
head, while I have examples from Geraldton, W.A., in which the thorax and 
underside are bronze (not coppery). C. aurulenta Kerr, is doubtfully distinct 
from this. 

Bicolor C. & G. is readily determined in a widely oval species well known 
from Western Australia, but which I have taken rarely in New South Wales, 

* Ethon breve, n.sp. 



162 REVISIOX OF THE GEXERA ETHON', CISSEIS AND THEIR ALLIES, 

and occurs in Victoria and South Australia. It varies in size from 8 to 14 ]nm. 
long, while a dwarf in the British Museum Coll. is only 5 mm. long. 

Marmorea C & G. is the species better known as acuducta Kirby, which has 
many synonyms and a wide distribution; probably the commonest species in 
Eastern Australia. Kerremans erroneously placed marmorata C. & G. under 
acuducta in the Genera Insectorum, instead of marmorea. 

MacuJata C. & G. is a small, narrow species, common, especially in Victoria 
and Tasmania, on Leptospermum. The white pubescence is readily rubbed from 
this otherwise bronze species, so that it rarely appears as in the figure of the 
Monograph; except as to the four apical spots (if Kerremans identification and 
my own are correct). I have not seen the type. 

In 1832 Hope redescribed 12-maculata F. under the name xantJwsticta. In 
1836 he published irrorata, lata and aenea (of which the last two = acuducta 
Kirby). In 1845 he added roseocuprea, signaticollis , gouldii, cupreicolUs (with 
its female form aeneicollis) , besides again repeating 12-maculata as 14-notata. 
Thus of 10 species, only 4 are to-day valid. I have examined all his types. 

In 1848 Germar described four species, notulata, chrysopygia, nuheculosa 
and chaUoptera. Of these, the first has been identified by Blackburn — ^whose 
authority as a long resident in the Adelaide District, and as a keen and able 
entomologist, carries weight — and I have followed him in this. The species is 
widel}" distributed throughout South Australia, Victoria and New South Wales. 

For chrysopygia, see under Neospades. 

Nuheculosa and chalcoptera are the respective female and male of the same 
species, the thorax of the male being metallic green; that of the female bronze. 
Analogous sexual colouration occurs at least in cupripennis Bois., marmorata 
C. & G. and, probably, in other species. 

In 1868 Saunders described similis and suturalis. The latter is referred be- 
low to Hypocisseis, the former sunk as a synonym. 

Maeleay (1872 and 1888) described eight species, of which latipennis is a 
Hypoeisseis, dimidiata, apicalis and purpureotincta are forms of the varicoloured 
chrysopygia Germ., impressicoUis and nigripennis are, I think, synonyms of 
roseocuprea Hope and Neospades simplex Blaekb. respectively, thus leaving only 
two valid species, viridiaurea and fulgidicollis , of which the former is a Neospades. 

In 1876 Gestro described (Ann. Mus. Gen., viii.) two species albertisii and 
cuprifera from Cape York (omitted by Kerremans from Gen. Ins.) ; the former 
of these is almost certainly synonymous with alho-sparsa C. & G., the latter is a 
Neospades of unusual beauty that occurs in Northern and North-West Australia. 

In 1879 Thomson published descriptions of thirteen species, of which seven 
are probably valid. I have not, however, seen his types. 

Regalis is, I think, an elongate species near leucosticta Kirby and fulgidicollis 
Mael., much less convex than the former, with a more sparse puncturation— the 
elytra violet-purple with irregular white pubescent spots. In Macleay's species, 
the form is generally less elongate and wider, with the pubescent spots definite 
in number and position. Opima, rugiceps, suhcarinifrons, uniformis and minu- 
tissima will be found in my tabulation and notes subjoined to this. Atro-violacea, 
as determined by Kerremans, varies from the description of that species in 
smaller size. Semiscabrosa, 1 have not been able to identify — it is probably one 
of the many forms of Neospades chrysopygia Germ. Viridicnllis is, I consider, 
the male of marmorata C. & G., but I have no decisive field e\ndence of this. The 
remaining four are obvious synonyms (see heloic). 



BY II. J. CARTEK. 1(J3 

Between 1887-1891 Blackburn described 13 species, including Neospades 
simplex and N. lateralis; besides Hypocisseis (Coraebus) pilosicollis. Of Black- 
burn's species, hella is probably only a form of N. simplex; dispar = roseoeuprea 
Hope^ verna = ■westicoodi C. & G. while constricta and lindi are, I think, the 
respective sexes of the same species. This leaves nine valid species. 

In 1898 and 1902 Kerremans described one Ethon and 41 species of Cjsseis 
from Australia (including two species of Neospades) . I have seen all his types. 
E. diver sum is an obvious synonym of E. fissieeps Kirby, of which the type is 
also in the British Museum. Of the 41 species, I can only recognise 12 as dis- 
tinct from previously described species. These are scahrosula, aurulenta, nitidi- 
collis, riridicepSj vicina, callaris, cyanura, nigro-aenea, viridis, fossicollis, rubi- 
cunda and puella, of Avhich aurulenta is doubtfully separable from leucosticta 
Kirby, collaris is doubtfullj^ distinct from vicina, while nigroaenea and viridis are 
referred to Neospades. The synonymy of the remaining 29 species is stated be- 
low. Many of the types are uniques — often mere variations from small species 
that are difficult of distinction. 

One other Australian Cisseis has been described by Thery as C. terrae- 
reginae. A specimen so labelled by Dr. Obenberger has been sent me, and 
corresponds to the description. It is a Neospades very near to, but distinguished 
from, N. viridaurea Mael. {See tabulation below). 

Buprestis lapidosa W.S. Macl.- — This species is wrongly placed amongst 
Cisseis in the Genera Insectorum, and other catalogues. The type is in the 
Macleay Museum, Sj^dney. It cannot be referred to any existing Australian 
genus, and is probably a misplaced exotic or, as in the case of B. au^rulenta F., a 
species bred in Australia from imported timber. Writing from memory in 
London, it is near the American genus Dicerca. 

The following tables will show my present view of the Genera Ethon, Cisseis,. 
Neospades, Hypocisseis and Alcinous. 

The following tabulation, it is hoped, will help to determine the species. 

Cisseis. Sect. i. 
Elytra clearly impressed with white pubescent spots. 

1 6 Ground colour of elytra dark blue. 

2 4 Pronotum blue, with pubescent vitta above lateral carinia. 

3 Disc without foveate impressions (8-10 x 4 mm.) 12-maculata F. 

4 Disc with two foveate impressions (5-7 x 2-24 mm.) elongatula Blkb. 

5 7 Pronotum coppery or brassy green, disc with pubescent foveate impressions, 

6 Pronotum (except on foveae) and underside glabrous. . . stigmata C. & G. 

7 Pronotum with inter-carinal space also sides of abdomen pubescent 

signaticollis Hope. 

8 Pronotum coppery (c?) or bluish (5) without pubescence, size small (5-7 
mm. 1.) cupreicollis Hope. 

9 15 Ground colour of elytra purple — sometimes blue (In this case distinguished 

from preceding by the coarsely punctate pronotum). 

10 12 Elytral spots indefinite in number, irregularly spaced. 

11 Form robust and convex— pronotum densely and coarsely punctate 

leucosticta Kirby. 

12 Form subdepressed, narrower than 11, pronotum sparsely punctate 

regalis Thorns. 

13 15 Elytral spots definite in number, eight arranged in a circle. 

14 Form elongate, disc of pronotum subsparsely punctate. . .fulgidicolUs Macl. 

15 Form widely ovate, pronotum finely rugose-punctate hicolor C. & G. 

16 18 Upper surface golden (pronotum sometimes green in anrtdenid). 



164 REVISION OF THE GENERA ETHON^ CISSEIS AND THEIR ALLIES^ 

17 Elytra multi-maculate aurulenta Kerr. 

18 Elytra with eight spots, definitely arranged alho-sparsa C. & Gr. 

19 Pronotum fiery copper, elytra brownish; the spots less defined than in 18. 

inflammata, n.sp. 

C. aurulenta Kerr, is possibly a small var. of leucosticta Kirby, a very vari- 
able species, widelj^ distributed throughout Australia. 

In the Monograph of Cast, et Gory, the figure of alhosparsa shows only six 
spots. Fresh examples, however, show eight. 

N.B. — The very distinct species 11, 12 and 14 (supra) are much confused in 
eoUeetions. 

C. cupreicollis Hope, is extremely like Neospades simplex Blkb., but has 
quite different tarsal claws, besides being shorter and less cylindric in form. 

C. elongatula Blackb. — The type of this, in the South Australian Museum, 
is a badly abraded example that makes identification dif&cult. There are, how- 
ever, so few Cisseis that combine the pubescent lateral vitta of pronotum with 
spotted elytra, that I consider three fresh examples in the same collection as 
conspecific. In these the elytra are clearly blue, with the spots arranged very 
much as in 12-maculata, showing also (in two examples) smaller, scattered, 
pubescent spots. 
Synonymy. 

(a) duodecemmaculata F. = 12-guttata Boisd. = 14-notata Hope = xantho- 
sticta Hope = pustulata Thoms. 

(&) leucosticta Kirby = stellulata Dalm. = fulgidifrons Kerr. 

(c) cupreicollis Hope = aeneicollis Hope (d*) = morosa Kerr. 

[d) alhosparsa C «& G. = albertisii Gestro = cupriventris Kerr. 

(a) Occurs from East to West of the Continent, breeding in, and found on 
leaves of Xanthorrhoea plants. 

(&) is perhaps the most widely distributed Cisseis in Australia, found in 
every state on Acacia foliage; I have examples from Geraldton, W.A., in which 
the pronotum and underside are bronze (not coppery), with the upper surface 
more coarsely punctured than usual. The type of fulgidifrons is only a small 
male of typical form and colour. 

(c) The type of morosa, is an abraded and discoloured cupreicollis Hope 
(var. aeneicollis Hope). 

Cisseis. Sect. ii. 

Elytra vagaiely impressed with white pubescence. 

Group A. Pubescence, in fresh examples, forming more or less circular 
spots. 

1 4 Size large (10-13 mm.' ), pronotum and underside sexually coloured (J 

green or bronze, ? more or less concolorous, with elytra rarely green). 

2 Form parallel, elongate — bronze or bronze green marmorata C. & G. 

3 Form widely oval, sides of pronotum well rounded, colour purpk, or 
purple bronze nubeculosa Germ. 

4 Form as in 3 but more acuminate behind, sides of pronotum nearly straight 
on basal half — colour dark bronze opima Thoms. 

.5 9 Form elongate-ovate — subparallel. 

6 Size medium (8-9 mm. ), head usually bilobed, pronotum, base and jipex 
of elytra bright purple, rest of elytra violaceous rugiceps Thoms. 

7 Size small (5^-7 mm. ). Pronotum and underside coppery bronze, elytra 
brilliant purple or violet tyrrhena, n.sp. 



BY H. J. OARTER. 165 

8 More or less concolorous, bronze, spots, except apical 4, very vague 

maculata C. & G. 

9 Pronotum brassy, or coppery green, elytra blue-black. . . nitidicollis Kerr. 
Notes on the above. — C. marmorata C. & Gr. is the large well-known species 

not uncommon in New South Wales and Victoria, of which the male is narrower 
and with metallic green or bronze pronotum and underside, described as vlridi- 
collis Thorns., and of which I have elsewhere described a bright green variety. 
The pubescence is often abraded — and obscure as in all the above. 

Nubeculosa Germ, is the female, chalcoptera Germ, the male of the same 
species — as recorded by Blackburn. The male is not unlike, when the pubescence 
is absent, a large sized cupripennis Boisd. It is common in South Australia and 
Western Victoria. 

Opima Thorns, is more triangular and of darker colour than nubeculosa, 
with a very differently shaped pronotum, wide but straight-sided behind, roundly 
narrowed in front. The males are similar in form but smaller. This species, 
apparently, only occurs in Western Australia. 

Rugiceps Thorns, is noteworthy for its strongly excavate head, which, to- 
gether with the pronotum, is unusually strongly sculptured — the sculpturing of 
the pronotum consisting chiefly of undulate transverse ridges. The purple colour 
carried over to the base of elytra is noteworthy, as also the same colour on apex, 
together with its green abdomen. 

C. maculata C. & G.^ — I have accepted the determination of Tasmanian ex- 
amples in the British Museum for this species, of which I have been unable to 
find the type; the pubescence is generally very vague, except as to the apical 
four spots. It is also the species described by Kerremans as tasmanica, indis- 
tinguishable from pauperula Kerr, (per type). 

G. nitidicollis Kerr, is very like ignicollis Kerr. (= Neospades simplex Blkb.), 
but the pronotum is much wider, and the claws those of Cisseis. 

Synonymy. — (a) marmorata C. & G. = simnlis Saund. = irrorata Hope 
(nee C. & G.) = viridicollis Thoms. = aenea Kerr. 

(&) nubeculosa Germ. = chalcoptera Germ. 

(c) maculata C. & G. = tasmanica Kerr. = pauperula Kerr. 

Group B. Pubescence, in fresh examples, not in spots; more or less marbled. 

1 26 Head not, or very feebly excavate between eyes. 

2 4 Size large (12-14 mm. long). 

3 Elongate oboviate, prothorax wider than base of elytra.. . . laticollis, n.sp. 

4 Ovate elliptic, prothorax narrower than base of elytra. , . elliptica, n.sp. 

5 9 Size smaller, (10 mm. long or less). 

6 Whole surface green, elytra coppery green ; form like opima. ovalis, n.sp. 

7 9 Upper surface more or less bronze. 

8 Rather widely oval, elytra nitid (8-10 mm. long) acuducta Kirby. 

9 Niarrowly oval, elytra with opaque subcyaneous patch. (6-7 mm. 1.) 

scabrosula Kerr. 

10 12 Size small (4-5 mm. long). 

11 Prothorax narrowing to apex, lateral carina not evident from above; 
tinged cyaneous parva Blaekb. 

12 Prothorax normally arcuate, lateral carina evident from above 

viridiceps Kerr. 

13 22 Pronotum and elytra differently coloured, the former more or less 

brilliantly metallic. 



166 REVISION OF THE GENERA ETHON^ CISSEIS AND THEIR ALLIES^ 

14 Form oval, pronotum golden copper, elytra amethyst blue, pulchella, n.sp. 

15 25 Form narrowly ovate. 

16 18 Elytra blue black. 

17 Pronotum bright bronze. atroviolacea Kerr. (? Thorns.) 

18 Pronotum metallic green vicina Kerr. ; ? collaris Kerr. 

19 Elytra violaceous, pronotum igneous purple, this colour more or less in- 
vading basal part of elytra notulata Germ.; ? violacea Kerr.; 

? nigrita Kerr. 

20 22 Elytra purple, pronotum bronze or coppery. 

21 Pubescence in scattered patches over whole elytra obscura Blaekb. 

22 Pubescence confined to apiaal areas rubicunda Kerr. 

23 25 Upper surface metallic green. 

24 Pronotum with large, deep fovea on each side fossicolUs Kerr. 

25 Pronotum without such foveae westwoodi C. & Gr. 

26 Form cylindi'ic, upper surface blue-black. cyanura Kerr. 

27 Head strongly excavated between eyes, pubescence scattered over whole 
elytra py g maea Blaekh. 

28 Smaller and more parallel than preceding puella Kerr. 

Semi-scabrosa Thorns, has been omitted from the above as unknown to me. 

It is possibly a variety of chrysopygia Germ. 

The species, especially the smaller ones (the greater number), of this section 
are very difficult to separate. Little is known of their sexual relation, and there 
is often a strong sexual colouration in the genus; while old, badly abraded or 
stained examples often present a widely different appearance from fresh ex- 
amples. Where more than one name appears under a number in the above 
tabulation it does not necessarily imply synonymy — but that the species are so 
close as to prohibit theii' clear tabulation by definite characters. 

Synonymy. — (a) acuducta Kirby = marmorea C. & G. = lata Hope = 
aenea Hope ^^ cuprifrons. Kerr. = laeta Ken*. 

(b) collaris Kerr. = ornata Kerr. 

The types are identical — vicina is doubtfully separated from these by its 
larger size, but an examination of a longer series is necessary. 

(c) notulata Germ. — Adopting Blackburn's determination of this South Aus- 
tralian species I consider that inops Kerr, and semiobscura Kerr, are identical 
with Germar's species, while violacea can only be distinguished by the fact that 
the elytra are wholly violet — whereas in the other three the basal area is more 
or less bright purple. Nigrita is, I think, but a dark form of the preceding. 
Thus notulata Germ. = inops Kerr. = semiobscura Kerr. = ? violacea KeiT. = 
? nigrita Kerr. 

(d) obscura Blaekb. = umdulata Kerr. = purpurea Kerr. 

(e) rubicunda Kerr. = modesta Kerr. 

Only slight colour difference separates the types. A western species. 

(/) parva Blaekb. = simplex Kerr. 

(g) viridiceps Kerr. = oblonga Kerr. 

(h) westwoodi C & G. = verna Blaekb. = viridana Kerr. = theryi Kerr, 

Occurs in Victoria, Tasmania and South Australia. 

(i) puella Kerr. = curta Kerr. I am doubtful as to the identity of this 
with pygmaea Blaekb. Botli have the Ethon-like head, and both occur in New 
South Wales, but puella is smaller and more parallel than pygmaea, while the 
less amount of pubescence on Kerremans' type may be accounted for by abrasion. 



4 


12 


5 


7 


6 




7 




8 


12 


9 


11 


10 




11 




12 





BY H. J. CARTER. 167 

CissEis. Sect. iii. 
Elytra without pubescent impressions. 

1 3 Forehead with a medial carina. 

2 Pronotum bronze, elytra violaceous, apex and apical margins coppery 
(8-9 mm. long) careniceps, n.sp. 

3 Upper surface purple or coppery bronze, margins of pronotum tinged green 

(5-7 mm. long) subcarenifrons Thorns.; ? oceidentalis Blackb. 

Foreheiad without carina. 

Upper surface dark. 

Pronotum abruptly constricted at base, elytra purple bronze, margins 

coppery constricta Blkb. 

Sides of pronotum nearly straight perplexa Blkb. 

Upper surface brilliant metallic. 
Sides of pronotum lightly rounded. 

Whole surface coppery bronze (6-7 mm. 1.) roseo-cuprea Hope. 

Whole surface green, or bronze-green (3i-4 mm. 1.). .. minutissima Thorns. 
Sides of pronottim obliquely narrowed from base to apex, colour uniform 
green (5-6 mm. 1.) uniformis Thorns. 

Synonymy. — (a) subcarenifrons Thorns. = cincta Kerr. = (?) var. ocei- 
dentalis Blkb. 

Thomson's description corresponds with many examples from King George's 
Sound, that are certainly cincta Kerr. (The type of oceidentalis is unique, I 
have been unable to match it. Smaller than Thomson's species, 5 mm. instead 
of 7 mm. , it is also of brighter colour, with & slight difference in the form of 
the lateral carina but I think these are conspecific). 

(b) constricta Blkb. = lindi Blkb. The former is the male, the latter the 
female, of the same species. Both forms occur in the British Museum Coll. from 
N.W. Aust., the male, as is frequent in the genus, having a bright metallic prono- 
tum and head. 

(c) roseocuprea Hope = impressicoTlis Macl. = dispar Blkb. == cuprea 
Kerr. = fairmairei Kerr. 

I have seen the types of all of these and find the first four identical. The 
last is, I consider, a variety in which the head and pronotum are greenish, the 
sides of the latter slightly wider than usual. A common species in Victoria and 
South Australia. 

(d) uniformis Thorns. = coraeboides Kerr. The dimensions were omitted in 
Thomson's casual description, but I have little doubt as to their identity. The 
examples examined are from Victoria and Tasmania. 

Perplexa Blkb. has the claws of a typical Cisseis though like Neospades 
nigroaenea Kerr, in size and form. The colour of the elytra is quite different, 
being dark violet, while the punctures on both pronotum and elytra are much 
more lightly impressed than in nigroaenea. This is a Western species. 

C. minutissima Thoms. — I have several examples that correspond with Thom- 
son's description of this species from Perth (J. Clark) and Pinjarra, W.A. (A. 
M. Lea). It was described as from South Australia. 

Cisseis inplammata^ n.sp. (Text-fig. 2.) 

Elongate ovate; head (of d*) gTeen, (of ?) fiery copper, pronotum, scutellum, 
underside and antennae fiery red copper, elytra brownish copper with green or 
blue tints near base, ruddy copper near apex, with some vaguely defined' spots 
smaller than, but placed as, in albo-sparsa C. & Gr. 



168 REVISION OF THE GENERA ETHON^ CISSEIS AND THEIR ALLIES, 

Read nearly flat, with feeble longitudinal impression, strongly, not densely, 
punctate. 

Prothorax: Apex nearly straight, base lightly bisinuate, sides (seen fiom 
above) obliquely narrowed from base to apex — the lateral margin (limited by 
lower carina) rather widely rounded in a vertical plane, anterior angles obtuse, 
hind acute; upper carina not nearly extending to apex, disc without a sign of 
medial line, sparsely and finely punctate in middle, becoming transverse laterally. 

Scutellum large, nitid and transverse, triangularly extended behind. Elytra 
widened at shoulder, rather strongly narrowed at apex; apices separately rounded 
and comparatively coarsely serrate, disc more distinctly punctate than usual, 
punctures large on basal half, fine and close near, apex — the eight pubescent spots 
vagTiely marked (in one example six spots only can be discovered). Dim. 8-9 x 
2i-3 mm. 

Hal). — Queensland: Johnstone River (H. W. Brown), Endeavour River 
(Melbourne Mus.), Cairns (Coll. Lea). 

Six examples, three of each sex, examined. It is nearest albo-sparsa C. & G. 
and tyrrlvena, n.sp. but unlike either. The coppery pronotum and underside are 
as in Melobasis pyritosa Hope. The colour of the elytra is elusive and variable with 
the light on it, that of the male type being suffused with blue. 

Types in Coll. Carter. 

CissEis MARMORATA C, & Gr. var. PRASiNA, n.var. 

Two examples (male) in the British Museum and two in the South Austra- 
lian Museum (Blackburn Coll.) are structurally so close to similis Saund. that 
I consider them colour varieties of that species, of which the males, as in many 
others, have the head, thorax and underside green or bronze-green. In those 
under notice the whole surface is metallic green, in one example peacock blue- 
green. Dim. 14-16 X SJ mm. 

Hab.— N.S. Wales (Saunders Coll. Brit. Mus.). 

C. viridicollis Thorns, has been already noted as probably the male of mur- 
morata C. & Gr. 

CissEis TYRRHENA, n.sp. (Text-fig. 3.) 

Elongate ovate. Head, pronotum, underside and appendages coppery bronze 
(head sometimes green) elytra brilliant purple or violet, with six not very 
clearly defined spots in a circular formation (as in albo-sparsa C. & Gr.), i.e., 
two near apex, two near sides, and two near suture near middle of elytra. 

Head canaliculate and depressed in middle, finely and densely punctate. 

Prothorax: Apex and base bisinuate, sides arcuately narrowed to apex, an- 
terior angle obtuse, posterior subrectangular, upper carina not extending to 
apex; subconeentrically punctate, with a foveate impression on each side near 
base. Scutellum transversely oval. Elytra rather narrowly oblong, covered with 
scale-like punctures, apices finely rounded, minutely serrulate. Sternum coarsely, 
abdomen very finely punctate, the seg-ments of latter with albopubescent spots at 
sides (in fresh examples). Dim. 5.5-7 x 2-3 mm. 

Hub. — N.W. Australia (DuBoulay and Saunders Coll. Brit. Mus.), Kala- 
mimda (H. M. Giles), Bunbury and Perth (J. Clark), Pinjarra (A. M. Lea). 

Many examples show a distinct species of narrowly oblong form, and of 
brilliant colours, somewhat suggesting nitida Kerr. The elytral pubescent spots 



RY H. J. CARTER. 



169 



can just be made out in some eases. Variations without the pubescent spots, 
sometimes with more obscure pronotum are in the British Museum, labelled Al- 
bany. The broader examples, where examined, are female and these in general 
show the maculae more clearly. Type in Coll. Carter. 

CissBis LATicOLLis, n.sp. (Text-fig. 4.) 

Elongate, oblong-obovate, subopaque bronze-violet above, pygidium tinged 
cyaneous, elytra with a vague, pubescent, fasciate impression near apex, and 
some pubescence near middle and sides; underside dull coppery bronze, often 
with a violet or bluish tinge — legs and tarsi following the colour of underside — 
antennae bronze. 

Head coarsely punctate, sparsely pubescent with a large, oval medial im- 
pression. 

Prothorax: Apex arcuate, base strongly bisinuate, sides very widely rounded, 
greatest width in front of middle, apex as wide as base, anterior angles (from 
above) acute, posterior obtuse, the two lateral carinae widely separated, sinuately 




„Jw-I 



Text-fig. 2. Cisseis inflammata, n.s-p. Text-fig. 3. C. tyrrhena, n.sp. Text-fig. 4. 
C. laticollis, n.sp. Text-fig. 5. C. elliptica, n.sp. Text-fig. 6, C. ovalis, n.sp. 
Text-fig. 7. C. pulchella, n.sp. Text-fig. 8. C. careniceps, n.sp. 



parallel till near base; a large fovea near apex, surface closely and finely, trans- 
versely rugose except a wide, nearly smooth, medial space; fresh examples with a 
fine grey pubescence scattered over greater part. Scuteilum very transverse. 
Elytra wider than prothorax at base and three and a third times as long, com- 
pressed near middle, enlarged behind; apices widely and minutely serrulate, these 
not covering pygidium and coarsely punctate; disc covered with scale-like punc- 
tures (much closer than in similis Saund.) these coarser towards apex; underside 
minutely punctate, with a sparse short pubescence. Dim. 11-14 x 4-5 mm. 

Hah. — Queensland: Wide Bay (British and Australian Mus.), Brisbane 
(Queensland and S. Aust. Mus.). 

A curious, irregular elongate species that can only be confused with similis 
Saund., from which it is clearly separated as follows : 



170 REVISION OF THE GElSrERA ETHON^ CISSEIS AND THEIR ALLIES; 

C. similis. C. laticollis. 

Prothorax widest behind middle, sides widest before middle, sides strongly 

lightly rounded. rounded. 

Elytra nitid, oblong-ovate. sub-opaque, oblong obovate. 

Type in British Mviseum. 

A male example in the Queensland Museum is much smaller (9 x 3.3 mm.) 
and has the head, pronotum and underside gTeen or bronze green. 

Var. CYANEOPYGA^ n.vax. 

Two examples from Western Australia (Lake Austin — H. W. Brown, in 
Coll. Carter) and British Museum, differ from the typical laticollis in having the 
prothorax widest at middle, the elytra less- Avidened behind, the pygidium bright 
blue, or blue-green. In the British Museum example the pronotum is also suffused 
with blue-green. This may prove, with the study of further material, to be distinct. 

CisSEis ELLiPTiCA_, n.sp. (Text-fig. 5.) 

Elongate elliptic, bronze or violet bronze, moderately nitid. The pronotal 
depressions pubescent as well as irregular pubescence on elytra, underside 
strongly pubescent, abdomen (in fresh examples) covered with white flocculenee. 

Head rather flat, finely channelled and pubescent, vertex coarsely punctate. 

Prothorax: Apex moderately, base strongly bisinuate, sides nearly straight, 
obliquely widening from apex to base, lateral carinae not parallel, the upper 
more sinuous than lower and continuous to apex, the lower unseen from above; 
all angles acute, the posterior wider; disc Avith two longitudinal depressions, one 
interrupted, irregular and wide on each side of middle, the other narrower inside 
apparent margin; finely sparsely punctate at middle, coarsely and transversely 
rugose towards sides. Scutellum transversely oval, punctate. Elytra lightly en- 
larged at shoulders, elliptically narrowed behind, apices separately rounded, hind 
margins strongly serrated, surface covered with scaly punctures, somewhat as in 
similis Saund., but becoming denser and finer towards apex; near base some 
longitudinal depressions continuous with those on pronotum. Underside, where 
not obscured by pubescence, coarsely punctate. Dim. 14 x 5 mm. 

Hah. — Western Australia: Cue and Tenindew^a (H. W. Brown). 

Two examples given me by their captor are abundantly distinct from lati- 
collis and similis by their convex elliptic form, oblique straight-sided prothorax 
Avith its unusually defined hind angles. Types in Coll. Carter. 

Yar. FRONTALIS. 

Two examples (the sexes) from N. Queensland (Kuranda — Dodd in Coll. 
Carter, and Herberton— Queensland Mus.) differ in having the head arcuately 
depressed between eyes, the sides of the arch raised above the level of the eyes 
and the underside nitid, lightly pilose, without flocculenee. 

CisSEis ovALis, n.sp. (Text-fig. 6.) 

Widely oval; head, pronotum, underside and legs green, elytra coppery 
green, vaguely impressed near apex, antennae and tarsi blue. 

Head arcuately impressed in middle, coarsely punctate. 

Prothorax: Apex and base bisinuate, the latter with medial lobe very wide, 
sides subparallel on basal half, thence arcuately narrowed to apex, space be- 
tween lateral carinae oval, upper carina not continuous to apex; two wide im- 
pressions near base, disc moderately convex, rather densely, subconcentricully 
rugose-punctate. Scutellum transversely oval. Elytra oval, apices widely, 



BY H, J. CARTER. 171 

scarcely separately rounded, margins near apex finely serrate, almost unil'oimly 
scalose punctate, a few vagme, sub-pubescent impressions near sides and apex, 
underside very nitid and densely punctate, sides of abdomen with pubescent im- 
pressions. Dim. 9x4 (vix) mm. 

Eab. — Western Australia (Du Boulay in British Museum). 

A single example, probably male, is quite unlike any of the described metallic 
green species. In shape near oiyima Thoms., in colour unlike any— though in this 
as also in elytra! sculpture it is near roseo-cuprea Hope. Type in British 
Museum. 

CiSSKis PULCHELLA, n.sp. (Text-flg. 7.) 

Oval. Head green or golden, pronotum and scutellum brilliant golden copper, 
elytra amethyst blue, with subfasciate white pubescence near apex; underside 
green, tinged with blue, sides of abdomen apparently non-pubescent ; legs, tarsi 
and greater part of antennae blue, basal joints of last coppery. 

Head lightly excavate, longitudinally canaliculate, densely and finely punc- 
tate. 

Prothorax: Apex and base bisinuate, narrowing from base to apex in slight 
curve, anterior angles obtuse, posterior acute from above; upper carina con- 
tinuous to apex, lower unseen from above, surface finely transversely sti-igose, 
transversely depressed near posterior angles. Scutellum large, transversely oval 
with triangular extension behind. Elytra with fine silky surface having minute 
scale-like divisions, scarcely punctate; apices separately rounded, their margins 
minutely serrulate, underside minutely punctate. Dim. 7 x 21 mm. 

Hah. — Queensland: Cooktown (H. Hacker, in Melbourne Mus.), Herbert 
River (Queensland Mus.). 

Three specimens of this brilliant little Buprestid are before me. In colour 
it is similar to Melohasis cyaneipennis Boh. (except for the elytral pubescence) 
and unlike any described species. Type in National Museum, Melbourne. 

CissEis CARExiCEPS, n.sp. (Text-fig. 8.) 

Elongate elliptic, navicular. Head, pronotum, underside and appendages 
coppery bronze, elytra varicoloured, tAvo examples chiefly violaceous, tinged 
cyaneous, the third example bluish-green; in all cases the suture (narrowly), 
apex and apical margins fiery coppery. 

Head rather fiat, with a dense longitudinal system of puncturatiou, and a 
well raised longitudinal carina in middle extending from l^etween the antenna! 
orbits to half-way between the eyes. 

Prothorax: Apex lightly, base strongly bisinuate, laterally convex, sides nearly 
straight, gently nai'rowed from base to apex; anterior angles obtuse, posterior 
acute (both from above) ; disc transversely striolate; rugose at sides; upper 
lateral carina unusually short, terminating some distance from apex. Scutellum 
transversely triangular. Elytra subparallel on basal half, finely narrowed at 
apex, apical mai-gins minutely serrulate, disc irregularly punctate, punctures 
dense near base; finely scalose-punctate near suture and apex, rather coarsely, 
transversely rugose-punctate on rest. Underside transversely striolate, with some 
faint lateral impressions on abdominal segments. Dim. 8-9 x 3 mm. 

7Jab.— Western Australia (Du Boulay) and N.S.W. (Saunder's Coll.). 

Three examples, labelled as above in the British Museum, show a species of 



172 



REVISION OF THE GENERA ETHON^ CISSEIS AND THEIR ALLIES^ 



iridescent pattern, remarkable for the well developed frontal carina and absence 
of elytral pubescence. 

C. suhcarenifrons Thorns., from Bang George's Sound, besides colour dif- 
ferences, has the following characters at variance with the above (1) "eaput- 
linea media longitudinale obsolete instructum"; (2) Prothorax "subglobuloso- 
cylindricus, nee transversus" ; (3) 7 mm. long. Type in British Museum. 



Neospades Black. 

Distinguished from Cisseis by (1) the lateral claws strongly bifid, (2) the 
basal joints strongly compressed, the posterior with basal joint scarcely longer 
than the second. The head is more or less bilobed, which is very exceptional in 
Cisseis, and the form, in general, parallel and convex. 

Neospades forms a convenient subdivision of the larger genus Cisseis. {See 
Text-figs. 9 and 10 for characteristic claws of Cisseis and Neospades. 

Table of species. 

1 7 Elytra chiefly golden, or green, with blue or purple marking, with or with- 

out pubescent impressions. 

2 4 Blue markings consisting of two spots and a post-medial fascia. 

3 Ground colour fiery copper, apex non pubescent. . . cruciata F. 

4 Ground colour golden green, apex with two pubescent spots 

cuprifera Gestro. 

6 7 Darker markings covering wide apical area with pubescent spots thereon. 
6 Darker markings continuous, along suture, to base, sides of pronotum 

nearly straight lateralis Blackb. 

7* Darker markings not continuous to base, sides of pronottmi rounded 

chrysopygia Germ. 

8 14 Elytral markings consisting of pubescent spots on a unicoloured ground. 

9 11 Ground colour of elytra dark, form subcylindric. 

10 Whole surface dark bronze, tinged violaceous gouldii Hope. 

11 Pronotum golden coppery, elytra dark blue simplex Blackb. 

12 14 Ground colour brilliant golden green. 

13 Elytra with two pubescent spots close to apex— the subapical spots clearly 
separate viridiaenea Macl. 

14 Elytra without pubescence close to apex — the subapical pubescence fcub- 
fasciate terrae-reginae Thery. 

15 17 Elytra without pubescence. 

16 Colour bronze-black nigro-aenea Kerr. 

17 Colour golden green viridis Kerr. 

Synonymy. — (a) lateralis Blkb. ^ splendida Kerr. 

(b) chrysopygia Germ. = dimidiata Macl. = apicalis Macl. 
= purpureo-tincta Macl. = semi-rugosa Thorns. = ? semi- 
scabrosa Thorns. 

(c) cuprifera Gestro = cuprifera Thorns. 

(d) simplex Blackb. = bella Blackb. = (?) nigripennis 
Macl. = ignicollis Kerr. 

(e) viridi-aurea Macl. = nitida Kerr. = viridicuprea Kerr. 




t Text-fig. 9. 
t Text-fig. 10. 



* N.Picta, n.sp., described below was received after this manuscript was com- 
plete, and would ioWov^ chrysopygia \n the table. 
+ Hind tarsal claw of Cisseis leucosticta Kirby. 
X Hind tarsal claw of Neospades lateralis Blackb. 



BY H. J. CARTER. 173 

(a) Tjrpes compared. The example of lateralis (eotype) in the Kerremans 
Coll. was abnormally coloured, and evidently misled him. 

(b). I have previously recorded the synonymy of Macleay's three species with 
semirugosa Thoms. I now accept Blackburn's determination of chrysopygia 
Germ, for examples in the South Australian Museum, which are clearly eon- 
specific. The species is common and widely spread in Queensland, South and 
Western Australia. 

Var: 

Two examples in the British Museum and one in the South Australian 
Museum have pubescent spots irregularly scattered over the apical half of elytra, 
besides the two regular spots usually found in this species. 

(c). Previously recorded by me (Arkiv for Zoologi, Band 13, No. 22, 1920). 

{d). 1 have above noted the strong similarity between N. simplex Blackb. 
and Cisseis cupreicollis Hope. The description as to colour "viridis, aureo- 
micans; elytris apacis obscure-aeneis" is misleading. The type has the pronotum 
and underside a brilliant golden green; the elytra are blue-black with pubescent 
spots (in fresh specimens in my collection, violet or clear blue as in C. 12- 
maculata F.) with coppery tints at suture and apex. When the pubescence is 
abraded, the forms like bella and ignicollis appear, and I believe nigripennis 
Macl. to be a discoloured example of the same. 

(e). I have closely compared the types, and find only slight differences of 
size and colour. 

Terrae-reginae Thery is very near viridi-aurea Macl. but has its pubescence 
differently placed, inter alia. 

Nigro-aenea Kerr, is differentiated from viridis Kerr., not only by colour, but 
by the close and deep sculpture of the former. Both are from Queensland. 

Neospades picta, n.sp. 

Oblong, lightly attenuate in front and behind; head pronotum, basal parts 
of elytra, underside and appendages nitid coppery bronze, apical two-thirds of 
elytra brilliant violet with preapical and medial pubescence. 

Head coarsely, not very closely punctate, forehead sub-bilobate (as in Cisseis 
pygmaeus Blackb.), eyes large and prominent. 

Prothorax transverse, apex strongly produced in middle, base bisinuate, sides 
rather straight, the inferior carina evident from above, widening from base to 
near apex, thence arcuately narrowed; whole surface coarsely, concentrically 
rugose, about a smooth oval spot situated at the apex of the front medial pro- 
duction. Scutellum transverse with a triangular extension behind, coarsely punc- 
tate. Elytra of same width as prothorax, separately rounded at apex, the bronze 
area extending to basal third at suture, nearly half-way at sides, the violaceous 
area thus forming an oval extension near suture. Pubescent markings (1) a sub- 
fasciate preapical one, consisting of a rounded patch on each side of suture and a 
disconnected transverse one extending to sides, (2) a more vague medial pubescence 
evident on sides, subobsolete on disc. Surface coarsely sealose-punctate, the scales 
arranged in irregular transverse ridges, these coarser on bronze, lighter on violet 
areas, underside lightly punctate. Dim. 6x2 mm. 

Hah. — Queensland: Brisbane (G-. H. Hardy). 



174 REVISIOK^ OF THE GFA^EEA ETIION^ CISSEIS AND THEIR ALLIES^ 

Two examples from the Queensland Museum show the typical Neospades 
structure — cjdinclric form, lightly divided head and forked claws. In one example 
the head is more brilliantly metallic than the pronotum, with a tendencj^ to green 
at the sides. In my tabulation the species would follow chrysopygia Grerm., from 
which it differs in its smaller, especially narrower, form; in its less deeply rugose 
surface, besides colour. Type in the Queensland Museum. 

Hypocisseis Thoms. 
(= Maschalix Waterh. = Cisseoides Kerr.) 

1 10 Colour blue-black or black, with white pubescence. 

2 4 Form wide, obliquely narrowed in front and behind. 

3 Head widely excavated, crested near eyes latipennis Mad. 

4 Head lightly excavated, without frontal crests hrachyformis Deyr. 

5 10 Form elongate oblong. 

6 12 mm. long, head widely excavated, eyes bordered within by narrow 
ridge pilosicolUs Blackb. 

7 8-9 mm. long, head lightly excavated, front scarcely ridged 

suturalis Saund. 

8 10 Head bilobed (somewhat as in Ethon), frontal lobes extended in front of 

eyes. 

9 Sub-apical pubescence viague and straight cyanura Kerr. 

10 Sub-apical pubescence forming parallel zig-zags rrdnuta, n.sp. 

11 Colour bronze with blue markings ; head widely excavated and ridged. . . 

ornata, n.sp. 

Notes and Synonymy of the above. 

Hypocisseis Thoms. = Maschalix Waterh. == Cisseoides Kerr. 

Waterhouse seems to have been unaware of Macleay's species in describing 
Maschalix latipennis as well as of Thomson's genus erected for the reception of 
Cisseis latipennis Mael. Cisseoides is to my mind superfluous, its distinctions 
from Hypocisseis being slight and of doubtful value. Kerremans himself described 
aeneipes (infra) as a Hypocisseis. 

(a) H. (Cisseis) latipennis Mael. = cornuta Gestro. = (Maschalix) lati- 
pennis Waterh. = latieornis Thoms. 

(&) H. (Coraebus) pilosicolUs Blackb. = Cisseoides murina Kerr. 

(c) H. (Cisseis) suturalis Saund. = Coraehiis marmoratus Mael. = Cis- 
seoides alhopicta Kerr. = Hypocisseis aeneipes Kerr. 

(d) H. (Cisseoides) cyanura Kerr. = C. modesta Kerr. 

With regard to («), I have seen the types of Macleay and Waterhouse while 
the descriptions of cornuta and latieornis make this synonymy certain. 

(h) Types examined at different times by me — but I am satisfied as to their 
identity. 

(c) Examples compared witli Macleay's type are in my Collection — the other 
types have ))een closely compared. This species is widely distributed from 
Queensland to Western Australia, and variable in size, and sometimes in ap- 
jjearance through abrasion of its pubescence. 

(d) The two types sliow slight colour variations only. H. hrachyformis 
Deyr., described from Myzole, is also found in Queensland. I have two examples 
from Rockhampton taken by Mr. H. W. Brown. It differs from latipennis in 
smaller size, flatter head, absence of frontal crests, the pronotum not channelled,, 
its sides rounded, its legs dark inter alia. 



BY H. J. CARTER. 175 

Hypocisseis minuta^ n.sp. (Text-fig. 11.) 

Oblong ovate, blue-black; elj^tra with transverse undulate impressions of 
white pubescence, underside and appendages eyaneous. 

Read canaliculate and divided, somewhat as in EtJwn, but the bilobed front 
produced in front of the eyes and more or less clothed with coarse, yellow hair, 
surface at vertex nearly smooth or very finely rugose. 

Prothorax very transverse convex, apex and base bisinuate, sides widely 
rounded, the two cai'inae enclosing a small narrow area, the convex disc sur- 
rounded at base and sides by horizontal depression more clearly eyaneous than 
the rest — the finelj^ rugose sculpture largelj^ concealed by shagreen clothing. 
ScuteUum large, transversely oval. Elytra oval, rather widely rounded at apex, 
apices scarcely separately rounded, and minutely serrulate; on apical area two 
parallel zig-zag pubescent impressions, a third and fourth similar impressions, 
decreasingly defined in proportion to distance from apex, a fifth vaguely-seen 
lunate pubescence behind scutellum from shoulder to shoulder. Surface else- 
where finely rugose, abdomen with a fine, short pubescence, breast minutely punc- 
tate. Dim. 4-5 X 1.7-2 mm. 

Hah. — Queensland: Johnstone River (H. W. Brown), Cape York (Coll. 
Carter) . 

Four examples in the South Australian Museum, also 3 in Coll. Carter (one 
from C. York) show a species near cyanura Kerr. (= modesta Kerr.) of which 
I have seen the types. Minuta is disting-uished from this, inter alia, by (1) 
lobes of head being pressed closely together — in cyamira widely V-shaped, (2) 
different pattern of elytral impressions— straight in cyanura. Types in South 
Australian Museum. 

N.B. — There is only a slight colour difference between the unique types of 
cyanura and modesta — both from Gayndah. 

Hypocisseis oenata_, n.sp. (Text-fig. 12.) 

Oblong, bronze, subnitid with the following markings 
eyaneous : a transverse fascia at base of pronotum, elytra 
with post-humeral spot and preapical fascia, interrupted 
ij ^ at suture. 

12 Head: A deep concavity between eyes, the sides or 

Text-fig. 11. flanges of concavity jutting beyond the eyes, each flange 

Hypocisseis with a large fovea within, eyes rather flat; width of head 

minuta, n.sp. less than that of prothorax at apex. Prothorax convex 

Text-fig. 12. and uneven in surface, rather gibbous and subangulately 

H. ornata, n.sp. produced at apex in middle, base bisinuate, widest at 

middle, basal half subparallel, thence subangTilately narrowed to apex, upper 
carina very short and sinuous, widely separated from lower carina in front; ir- 
regulai'ly, transversely depressed near base, the depression more or less eyaneous. 
Elytra oblong-acuminate, obliquely narrowed behind, with four longitudinal cal- 
losities near base, one on each side of scutellum, and one on each shoulder, the 
latter connected with a little raised costa extending obliquely from shoulder to 
middle of elytra near apical declivity; a transverse gibbosity also behind scutel- 
lum. Surfa,ce closely and finely punctate, the eyaneous markings vagniely out- 
lined by white pubescence. Underside glabrous and more coarsely punctate. Bim. 
4^-6 X 2-24 mm. 




176 REVISION OF THE GENERA ETHON^ CISSEIS AND THEIR ALLIES. 

Hah. — South Australia: Lucindale (Feuerheerdt) ; Western Australia: 
Geraldton (W. D. Dodd). 

Eight examples of this pretty little insect examined. The areuately excised 
head is somewhat as in suturalis Saund., or (Coraehus) pilosicollis Blackb., but 
the excision is deeper and wider. In two examples the blue markings on thorax 
and side of elytra are not very clear. The species is a bridge between H. lati- 
pennis Mael. and suturalis Saund. Types in the South Australian Museum. 

Alcinous Deyr. 
Ann. Soc. Ent. Belg., 1864, p. 115; Kerr., ibid., 1898, p. 174. 

Synonymy.- — A. nodosus Kerr. = A. minor Kerr. 

I have examined the types and consider them mere variations of a species 
that I have taken in many localities of Eastern Australia. The genus is some- 
what intermediate between Cisseis and Agrilus, of shorter form than the latter, 
and readily distinguished from Cisseis by its irregular, nodulose surface. l^ike 
Cisseis it also occurs on foliage, chiefly Acacia — in the coastal brush areas of 
Queensland and New South Wales, but I have also taken an example at Yea, 
Victoria. 



177 



THE LIFE HISTORY OF MICBOCACEBYS TETBAGONA (HOOK.). 

By A. Anstruther Lawsoist, D.Sc, Professor of Botany, University of Sydney. 

(Plates xv.-xvi. and Thirty-four Text-figures.) 

[Read 30th May, 1923.] 

Microcachrys tetragona (Plates xv.-xvi.) is one of the rarest and at the same 
time one of the most interesting of the Coniferales. It is a low, creeping, pro- 
fusely branching shrub inhabiting the higher mountain tops of Central and Wes- 
tern Tasmania. It clings very closely to the windswept and often snow-covered 
rocks and, with its numerous small branches and closely imbricate scale-like leaves 
in four rows, it is a remarkable adaptation to high alpine conditions. Its habit 
and habitat, together with its crimson coloured succulent cones, present features 
quite unusual among gymnosperms. Its geographical distribution is limited to 
the higher elevations of the Tasmanian mountains. As a monotypic genus it was 
first described by Hooker in 1845. 

Its rare occurrence and the inaccessible regions it inhabits doubtless account 
for our lack of knowledge regarding its life-history. With the exception of a 
brief notes on the pollen by Thompson (1908 and 1909) we have no knowledge of 
the gametophytes of this interesting genus of the Podocarpineae. 

It seems very desirable that this and other little-known Australian Conifers 
should be investigated, with the hope of filling in the gaps in our knowledge of 
this great sub-division of the Gymnosperms. 

Being within fairly easy distance of Tasmania and the mountains inhabited 
by Microcachrys^ I have been enabled, as a result of a number of expeditions, to 
obtain a complete series of developmental stages of the gametophyte structures, 
fertilisation, and embryo of this rare genus. 

A very considerable amount of the material forming the basis of this in- 
vestigation was collected and fixed in the field. Some of it, however, was from 
time to time sent by post. In such cases the specimens were carefully packed in 
moist Sphagnum and reached my laboratory, generally within forty-eight hours, 
in the living condition. Upon its arrival, the specimens were immediately dis- 
sected and fixed in dilute ehrom-acetic acid and othei-wise prepared for imbedding 
in paraffin. Prom these two sotirces enough material was secured to show all 
the essential details of the life-history. 

The Male Gametophyte. 

Prom a study of the living microspores it became quite evident that the 
pollen is winged. A superficial observation under the microscope might lead one 
to the conclusion that there were two wings present. But a slight pressure on 



178 



THE LIFE-HISTORY OF MICROCACHRYS TETRAGONA^ 



the cover glass will cause the microspores to turn or roll and the presence of a 
third wing is easily demonstrated. From a study of a series of stained sections 
through the mierosporangium, the presence of three wings was abundantly con- 
firmed. These wings which are inflations of the spore-wall are arranged in a per- 
fectly symmetrical fashion, as indicated in Text-fig. 1. This figure represents a 
section of a microspore showing only one nucleus. The thickness of the spore- 
wall and the contained dense cytoplasm is indicated. I was unable to detect any 
noticeable difference in the size or shape of the wings. In many of the sections, 
portions of the wings were removed in the cutting, giving an unsymmetrical ap- 
pearance, but in the thousands of living microspores examined there was a re- 
markable uniformity. Text-fig. 1 may be taken as a very fair sample of the 
three inflations or wings. From living material collected in the field I have not 
found any microspores with more than three wings. In this connection Thomp- 
son (1909) has reported a slight variation. 






Text-fig. 1. — A section of a micrcspore showing one nucleus. The thick cell wall 

is dilated in three places, forming three wings of equal size and 

symmetrically arranged. Dec. 10th, 1921. 
Text-fig. 2. — A section of a microspore showing two of the wings; the wall of 

the body cell is clearly indicated as well as the tube nucleus. Dec. 

10th, 1921, 
Text-fig. 3.— A section of a microspore showing two prothallial cells, a tube 

nucleus; the generative cell has just divided to form the body cell 

and stalk nucleus. Dec. 10th, 1921. 
Text-fig. 4. — A section of a mature microspore at the time of polHnation. Here 

only two of the three wings are shown. Within the spore which 

has germinated there are two prothallial cells, a large central body 

cell, a stalk nucleus and a tube nucleus. 
(Figs. 1, 2, 3. 4. X 1000) 

In regard to the contents of the microspore, I find that the mature pollen 
grain at the time of pollination has within it three cells and two free nuclei. 



BY A. A. LAW SON. 



179 



Under favourable conditions these could be observed in the living state. Text-fig. 
4 may be taken as typically representing a median section of a microspore at 
this time. It will be observed that the three cells are separated from one another 
by thin but sharply defined cell walls. The three nuclei of these cells lie directly 
one after the other in a straight line and one of the free nuclei lies behind in the 
same line. The other free nucleus is invariably at one side. There are thus four 
nuclei in a row and one excentrieally situated. My interpretation of these struc- 
tures is as follows : — ^The two smaller cells nearest the spore-wall are the two pro- 
thallial cells, con'esponding to the two prothallial cells so characteristic of the 
Abietineous pollen (Coulter and Chamberlain, 1901, 1910). The large centrally- 
situated cell is the body-cell and the excentrically-situated free nucleus is the stalk 
nucleus. The remaining free nucleus is the tube nucleus. Although many hun- 
dreds of microspores were examined, both in the living condition and in the 




Text-fig. 5. — A section of the tip of the pollen tube showing the large spherical 
body cell and the stalk nucleus, and the tube nucleus suspended in 
the tube cytoplasm. The tube nucleus is just forward of the body 
cell. Feb. 6th, 1922. 

Text-fig. 6. — A shghtly older condition of the above. The body cell loses its 
spherical form and its nucleus prepares for mitosis. Feb. 6th, 1922. 

Text-fig. 7. — From a section of a pollen tube near the tip. The body cell has 
divided to form two sperm cells of approximately equal size. The 
stalk and tube nuclei are also shown. Feb. 6th, 1922. 
(Figs. 5, 6, 7. xlOOO) 

sections, I was unable to find a single instance of a microspore containing more 
than five nuclei, namely, those of the two prothallial cells, that of the body-ceU, 
the stalk nucleus and the tube nucleus. I am therefore unable to confirm Thomp- 
son's (1909) observations in this connection. Text-fig. 3 represents a stage 
slightly earlier than that represented in Text-fig. 4. Here the two prothallial 



180 IHE LIFE-HISTORY OF MICROCACHRYS TETRAGONA^ 

cells are fully organised, but the stalk and body cells are not. Evidently this is 
a stage where the generative cell has just divided and the wall separating the 
body cell from the other structures is not completed. Text-flg. 2 is taken from a 
section of a spore containing the tube nucleus and the body cell. It is intended 
to show how clearly defined is the wall of the body cell from this view. 

Text-fig. 22 shows the manner in which the microspores are deposited— 
evidently driven by the wind — in the micropyle. Here these microspores may be 
seen with their winged inflations and their nucleai- contents. The reflexed ex- 
tension of the integument is very characteristic and probably assists in the reten- 
tion of the microspores. Upon reaching the apex of the megasporangium, the 
microspores continue their germination, by producing a pollen tube. In Text- 
fig. 25 one may observe how the pollen tube penetrates the nucellar tissue. At 
the apex of the megasporangium a sharply defined cushion of tissue which, in 
another investigation, I have referred to as the poUen cushion (Lawson, 1910) is 
organised. Through this the pollen tube descends, dissolving the nucellar tissue 
in its advance, carrying the body-cell and the tube and stalk nuclei at the tip. As 
the archegonial chamber is reached the tube dilates and spreads over the apical 
region of the prothallus. Text-figs. 26 and 27 show the dilated tip of the pollen 
tube with its contents. During the descent of the pollen tube, the body-cell • 
gradually grows until it is manj times its original size. It assumes an almost 
perfectly spherical form and its large, centrally-situated, deeply-staining nucleus 
is a conspicuous feature in the section. Text-flg. 5 represents the tip of the 
pollen tube at this time. The large body-cell is here seen suspended in the dense 
mass of cytoplasm. It has a very thin but well defined cell-wall and on either 
side of this are the stalk nucleus and the tube nucleus respectively. I can inter- 
pret these three stnietures in no other way than that they are the same three 
structures indicated in the microspore' (Text-flg. 4), the two prothallial cells 
having remained in the spore on the apex of the nueellus (Text-fig. 25). 

Text-fig. 6 represents a pollen tube slightly older. Here the body-cell has 
enlarged still more. It has lost its spherical form and its nucleus is preparing 
for mitosis. The stalk nucleus is seen at one side. 

The actual division of the body-cell was not observed, but in one instance the 
result of this mitosis was clearly seen. This is shown in Text-fig. 7. There are 
evidently two sperm cells formed, each with its own thin cell-wall. As these two 
sperm cells were pressing closely together it was difficult to say whether there 
was any difference in the size. Apparently they were approximately equal. 

The male gametophyte and details of spermatogenesis of Microcachrys, while 
essentially resembling Podocarpus, show some striking differences (Buiiingame, 
1908). For Podocarpus, Coker (1902) reports as many as six nuclei in the 
mature pollen grain. He states, however, that this is probably abnormal. It is 
to be noted also that in Podocarpus there is but one functional male cell. In 
Microcachrys the male gametes are probably both functional. 

The Female Gametophyte. 

The ovulate strobili are situated at the ends of short branches, and although; 
in the mature state, they are bright red in colour they are not conspicuous in the 
field, being more generally hidden by other sterile branches. A perfectly median 
section of a very young sporophyll is shown in Text-fig. 23. It will be seen that 
the bract grows nearly at right angles to the cone-axis for the greater part of its 
length, but at a point near its distal end it turns sharply in a vertical direction. 
At this point, on the adaxial face of the sporophyll, is situated the megasporan- 



BY A. A. LAWSON. 181 

g'ium with its integument, it being protected by the upward growth of the apex 
of the sporophyll Avhich is now recurved towards the axis. The scale at this 
early stage follows a course in its growth almost parallel with that of the distal 
end of the sporophyll. It should be pointed out that there is only one ovule to 
each bract and this latter, even at this early stage, as shown in Text-fig. 23, is 
recurved in its growth until it is directed in a line with its mieropyle towards 
the cone-axis. The extended mouth of the inner integniment projects for a con- 
siderable distance beyond the nueellus and recurves in a downward direction to- 
wai'ds the bract, forming quite a long tvibe-like mieropyle. Typical instances of 
this feature are illustrated in Text-figs. 22, 23 and 24. In Text-fig. 24 the single 
vascular strand may be seen supplying the sporophyll and branching into the 
ovule. The large, loose, thin-walled cells seen at the curvature of the sporophyll 
are an indication of the future succulent nature of this tissue. 

The conditions here described are those found at the time of pollination. The 
microspore cones are produced in great profusion and the resulting poUen pro- 
duced is very great indeed considering the size of these dwarf plants. The driv- 
ing winds in these exposed alpine regions cover the plants with pollen so that 
very few ovules escape its reception (Text-fig. 22). 

At this time the megaspore mother-cell makes its appearance. In the very 
early condition it is difficult to distinguish this from other nucellar ceils, but at a 
point in line with the base of the integument a single mother-cell becomes dis- 
tinctly' differentiated. It is larger, clearer and has a much more conspicuous 
nucleus than the neighbouring cells. The cells immediately surrounding it also 
undergo a change and become a well defined tapetum of two or three cells deep. 

The first appearance of the mother-cell is shown in Text-fig. 23; slightly 
older stages are indicated in Text-figs. 22 and 24 in which the tapetum is shown. 
A somewhat older stage is shown^in Text-fig. 21. This is from a transverse sec- 
tion to show the relationship of the integument with the nueellus. In the centre is 
a large mother-ceil preparing for reduction division and surrounded by three 
layere of tapetal cells. 

I was fortunate enough to secure an almost perfect series of stages in mega- 
sporogenesis. From the abundance of material at hand it seems quite clear that 
only one megaspore mother-cell is organised. A very typical appearance of this 
cell is shown in Text-fig. 8. The relative size of this cell with its large nucleus is 
very striking. It is many times greater than the largest of the surrounding 
tapetal cells. The cytoplasm of this mother-cell, while finely granular, is much 
more clear than the dense cytoplasm of the suiTounding tapetum and hence stands 
out in the greatest possible contrast. Although the actual spindle was not found, 
all the essential details of the sp orogenesis were observed. In Text-fig. 8 the 
nucleus is represented in the spireme state. The long thread-like chromosomes 
are clearly defined and are evidently split lengthwise. In Text-fig. 9 one sees the 
chromosomes actually uniting in pairs, which results in reduction. The number 
of the chromosomes seems to be twelve but, as the specimens examined in these 
stages were limited, jfurtber confirmation is necessary. Twelve were found in 
about twenty nuclei. There is no doubt, however, that this is a phase of the re- 
duction process. The result of this first meiotic division is shown in Text-fig. 10. 
Traces of the spindle fibrils are still seen stretching between the daughter-nuclei 
and a distinct cell-wall separates the latter from one another. This division is 
evidently immediately followed by another. The two daughter-nuclei divide, but 
apparently not simultaneously. Text-fig. 11 shows one of these divided and the 
other near the mieropilar end still vmdivided. In Text-fig. 12 both these nuclei 



182 



THE LIFE-HISTORY OF MICEOCAGHEYS TETRAGON A^ 



have divided so that the four megaspores are represented at this time by two 
cells each with two nuclei. At the completion of this second meiotic division, 
walls separate all the nuclei, so that there are four separate cells forming an axial 
row. Only one of these megaspores functions, the proximal one, that nearest the 
food supply. The other three degenerate and appear as flattened masses of 





e^& 



Text-fig. 8. 



Text-fig. 9.- 



Text-fig. 10. 



Text-fig. 11, 





12 •#! 



-From a section taken in a median line through the megasporangium. 
In the centre of the figure is the large megaspore mother cell sur- 
rounded by a well developed tapetum. Dec. 10th, 1921. 

-From a similar section to Text-fig. 8. Here the nucleus of the 
megaspore mother cell is in the first state of meiosis. The bivalent 
chromosomes are uniting in pairs. A well organised tapetum is also 
indicated. Dec. 10th, 1921. 

-A section to show the result of the first meiotic division of the 
megaspore mother cell. Two cells are now formed, each with its 
own nucleus, and separated from one another by a distinct mem- 
brane. Dec. 10th, 1921. 

-A similar section to show the result of the second meiotic division. 
Here one of the nuclei shown in Text-fig. 10 has divided. There 
are now one cell and two nuclei in the axial row. Dec. 10th, 1921. 



BY A. A. LAW SON. 183 

nuclear substance at the distal end of the functional spores. This latter has mean- 
time enlarged many times its original size. Text-fig. 13 is a typical example of 
numerous instances found of this condition. During its growth an immense 
vacuole is developed in the centre of the germinating megaspore and this forces 
the cytoplasm to occupy a parietal position. At this time the tapetum has reached 
its maximum development. These features are well illustrated in Text-fig. 13. 
The nucleus of the megaspore now immediately undergoes a series of free divisions 
and the nuclei thus formed take up positions in the parietal cytoplasm (Text-figs. 
13, 14, 15). This is the first instance in which the essential details of mega- 
sporogenesis have been reported for the Podocarpineae. In the near future I 
shall compare these stages with others which I have under investigation. This 
series of events leading to the mature functional megaspore is essentially the same 
as that in Sciadopitys (Lawson, 1910). With this development the tapetal cells 
undergo degeneration and their substance is evidently absorbed by the germinat- 
ing megaspore. As shown in Text-fig. 14 the tapetum has completely vanished 
and the inflated megaspore, with its numerous free nuclei and large central 
vacuole, presses and flattens out many of the nucellar cells with which it comes in 
contact (Text-fig. 15 and 16). So far there is no indication of a cellular pro- 
thallus, the young gametophyte dbnsisting of the large central vacuole enveloped 
by parietal cytoplasm and at least sixty-four free nuclei. But soon after this 
stage shown in Text-fig. 16, the mitoses result no longer in free nuclei. Each cell 
that is now formed is completely surrounded by a cell wall. So that, instead of 
parietal cytoplasm, there is a single row of narrow cells lining this cavity. Of 
all the Gymnosperms I have examined, I have not found such beautiful examples 
of this so-called "Alveolar" structure, first discovered by Sokolava (1880), and 
described by Arnoldi as "alveoli" (1900). 

These cells are apparently first formed at the distal end of the vacuole and 
gi-adually come to line the whole space occupied by the jDarietal cytoplasm. While 
the cells or "alveoli" are being formed in the proximal end the entire mass of the 
parietal cytoplasm is not included within their walls. Many preparations showed 
considerable masses of this outside the cell-walls as for instance shown in Text- 
fig. 18. This figure also shows the rapid elongation of these "alveoli." The 
rapid growth of these cells into long tubes, each at first with a single nucleus, is 
doubtless assisted by the absorption of the vacuolar sap. For this latter now be- 
comes less and less until the entire space it once occupied is encroached upon by 
the "alveoli." A very fair example of this condition is shown in Text-fig. 19. 
Here the primary cells of the young prothallus are seen to have elongated to such 
an extent that the vacuole has nearly vanished. The rate of growth of these 
primary cells is not uniform. Those at the distal end appear to have a gTeater 
growth than those at the opposite end. Eventually, however, as was observed in 
both longitudinal and transverse sections, the vacuole becomes entirely obliterated 



Text-fig. 12. — A similar section to show the second cell shown in Text -fig. 10 has 
divided. The separating walls have not yet been laid down. There 
are eventually four cells in the axial row. These are the mega- 
spores. Dec. 10th, 1921. 

Text-fig. 13. — From a similar section a few days older. The proximal cell of the 
axial row has grown to a great size and becomes the functional 
megaspore. The other three cells become aborted and appear as 
three nuclear masses pressed into the tissues of the tapetum which 
still persists. The nucleus of the megaspore has divided to form 
two free nuclei. Dec. 10th, 1921. 

(Figs. 8-13. x550) 



184 



THE LIFE-HISTORY OF MIOEOCACHRYS TETRAGONA, 



and then tlie mitoses following result always in cell-walls being formed. Oc- 
casionally it was observed that these primary prothallial cells had two or even 
three nuclei, but the resulting smaller cells have only one. The prothallial cel- 




Text-fig. 14. — A section through a median line of the megasporangium, at a much 
lower magnification than the preceding six figures. Here the mega- 
spore has continued its germination and its nucleus has undergone 
repeated free nuclear division. The tapetum is in a state of dis- 
organisation and rapidly being absorbed by the developing pro- 
thallus. Jan. 21st, 1922. 

Text-fig. 15. — A later stage of the same. Here the tapetum has completely 
vanished and the megaspore has increased to many times its 
original size. Free nuclear division has proceeded and the large 
clear central vacuole forces the cytoplasm into a parietal position 
and here the free nuclei take up their position. Jan. 21st, 1922. 



BY A. A. LAWSON. 185 

lular tissue is thus formed. The original line of cells set down in the early 
stages by the elongated primary protliallial cells is retained in the mature pro- 
thallus. Test-fig. 27 represents a longitudinal section of the upper half of the 
prothailus. Here, on either side of the archegonia, the curious rows of cells are 
those orginally laid down in the "alveolar" style. 

The archegonial initial was not observed. Many preparations showed the 
mature archegonia. Curiously enough these are developed close together in a 
single group after the manner of the Cupressineae (Lawson, 1907, Coker, 1903). 
A common nourishing jacket two or three cells deep envelopes the archegonia. 
There is likewise a single common archegonial chamber into which the tips of the 
pollen-tubes enter. Each archegonium is three or four times as long as broad. 
There are four neck cells and a large vacuole in the lower half of the cytoplasm. 
The egg nucleus lies between this vacuole and the neck cells. As the very young 
stages in the development of the archegonia were not found I was unable to de- 
monstrate the prepence of a ventral canal nucleus. But doubtless this is present 
as it is found in so many other Taxaceae. 

Among the Podoearpineae there seems to be a considerable variation in the 
number of archegonia. One species of Phyllocladus is reported to have but one 
(Kildahl, 1908), and there are as many as eleven in Podocarpvs (Coker, 1902). 
Five or six was the usual number met with in Microcachrys. A characteristic 
appearance of the archegonia in longitudinal section is shown in Text-fig. 27. 
There are three mature archegonia here shown crowded closely together and with 
their necks opening into a common archegonial cavity at the apex of the pro- 
thailus. These are ready for fertilisation. Text-fig. 28 is an enlarged detail of 
a typical archegonium to show not only the egg nucleus, vacuole and neck cells, 
but it .shoAvs very well the highly developed jacket cells in contact with it. The 
number of neck cells is four, forming a, single tier. This is in sharp contrast to 
other Podoearpineae, differing in this particular from Podocarpihs (Coker, 1902) 
and Phyllocladus (Young, 1910). 

A conspicuous feature of the mature gametophyte is the megaspore mem- 
brane. This is a thick tough membrane which completely envelopes the prothailus 
except at the apex in the region of the archegonial chamber. This membrane is 
so tough and thick that it interfered considerably with the fixation of the material 
and also Avith cutting sections with, the microtome. The difficulty of fixation was 
overcome by carefully cutting away the lower portion of the ovule in such a way 
as to make an exposed opening for the fixing fluid to enter. This megaspore 
membrane is quite as thick and conspicuous as that occurring in the Abietineae. 

Text-fig. 16. — The same as Text-fig. 15, but a more advanced stage. Jan. 21st, 
1922. 

Text-fig. 17. — The same, but showing that the free nuclear division has ceased, 
and cells are now formed occupying the position of the parietal 
cytoplasm. This immense clear central area of the vacuole is now 
fined with a layer of closely fitting cells whose long axes are directed 
towards the centre of the vacuole. Each cell has a single nucleus. 
Jan. 30th, 1922. 

Text-fig. 18.— A similar section to show a further development of these primary 
cells or alveoli of the gametophyte. These cells are elongating to- 
wards the centre, especially those at the distal end. Jan. 30th, 1922. 

Text-fig. 19.— A later stage of the same. The primary prothalfial cells or 
"alveoli" form a series of long tubes which in their growth encroach 
upon the vacuole. The latter is nearly obliterated. Jan. 30tb, 
1922. 

(Figs. 14-19. x200) . 



186 



THE LIFE-HISTORY OF MICROCACHRYS TETRAGOKA;, 



It is double-layered aud suberised. It is curious that the Taxineae and Podo- 
carpineae should differ so consistently in this detail. In the former group 
(Thompson, 1905, Lawson, 1909) it is quite thin and inconspicuous, while in the 
latter, that is, in Phylloclad^s, Dacrydium, Saxegothaea and Microcachrys it is 
weU developed. A high power magnification of this membrane is shown in Text- 
fig. 20. 

Fertilisation. 

By the time the archegonia are organised, the dilated tip of the pollen tube, 
has reached the archegonial chamber. The course of the tube and its dilated end 
are shown in Text-fig. 27. Here the tube is well into the archegonial chamber, 
and it contains the body cell and stalk and tube nuclei. These latter are usually 
in advance of the body cell. In Text-fig. 25, one may see how very large the tip 
of the tube becomes as it spreads out over the apex of the prothallus. In Text- 




Text-fig. 20.— A section of a portion of the mature prothallus showing the nature 
of the megaspore membrane. Feb. 6th, 1922. 

Text-fig. 21.— A transverse section through the basal region of the megasporan- 
gium. The outer mass of tissue is the integument. The dark line 
marks the separation of the integument from the nucellus. The 
large megaspore mother cell surrounded by a three-layered tapetum 
occupies the centre. Dec. 10th, 1921. 

(Figs. 20. x300; fig. 21. x 70) 

fig. 26 the nature of the l)ody cell close to the archegonial neck is well illustrated. 
The sperm cells, however, have not yet been formed. The conditions shown in 
Text-figs. 5, 6 and 7 were taken from similar positions, that is, in close proximity 
to the necks of the archegonia. The division of the body cell therefore does not 
take place until just before fertilisation. The result of this division is shown in 
Text-fig. 7 which we have already described above. In one instance the actual 
penetration of tlie male gamete into the archegonium was observed but its actual 
union with the e^g nucleus was not found. As the arfhcgonia form a single 



BY A. A. LAWSON. 



187 




Text-fig. 22. — A section taken in a median line through the megasporangium. The 
recurved tube of the integument is shown forming a micropyle con- 
taining three microspores. A megaspore mother cell surrounded by 
its tapetum is seen in the centre of the nucellus. Dec. 10th, 1922. 

Text-fig. 23. — A median section through a megasporophyll. The characteristic 
position of the megasporangium with its integuments is clearlv 
shown, as well as the characteristic position of the integument and 
the micropyle. This and the preceding figure were taken at the 
time of pollination. Dec. 10th, 1922. 

Text-fig. 24.- -A similar section of the megasporangium to show the characteristic 
appearance of the integument, micropyle, nucellus, megaspore 
mother cell and tapetum. Dec. 10th, 1922. 

(Fig. 22. x200; fig. 23. xlOO; fig. 24. x200) 



188 



THE LIFE-HISTORT OF MICEOCACHRYS TETEAGONA^ 




Text-fig. 25. — This figure represents a longitudinal section of the upper part of the 
megasporangium. It shows the v/inged microspores on the apex of 
the nucellus which is differentiated into a pollen cushion. The 
figure is to show the course of the pollen tubes through the nucellar 
tissue. Feb. 6th, 1922. 

Text-fig. 26. — This represents a .section through the apex of the prothallus. The 
tip of the pollen tube has discharged its contents into the arche- 
gonial chamber. Two archegonia are seen as well as the charac- 



BY A. A. LAWSON. 189 

gi-oup with a common jacket and a common arehegonial chamber, it seems highly 
probable that lioth male gametes, which appear to be ec|nal in size, are func- 
tional^fertilising different archegonia. In this respect Microcachrys differs 
from Phyllocladus and Torreya* in which there is a marked difference in the size 
of the two gamete?.. In the latter two genera, one of the two male gametes is 
aborted. In Cephalotaxus drupaoea (Lawson, 1907) they remain within the 
jacket of the mother cell until the contents of the pollen tube are discharged. It 
seems that in Microcachrys all of the archegonia were fertilised and this supports 
the view that both male cells in each tube are functional gametes. I am inclined 
to believe that the i-eduction of one of the gametes is an adai^tation to the manner 
in which the archegonia are grouped or have a common arehegonial chamber. 
"When the entire contents of a pollen tube are discharged into a single arche- 
gonium one should expect one of them to be functionless. If on the other hand 
the archegonia necks are so arranged that the two male gametes may enter dif- 
ferent but neighbouring archegonia one would expect them to be both functional. 
This is what appears to be the case in Microcachrys. This is very characteristic of 
the Cupressineae (Lawson, 1904, 1907, Coker, 1900). 

The Embryo. 

In regard to the embryo I unfortunately was unable to find the early free 
nuclear condition. The early stages in the development of the suspensor, however, 
were frequently met with. From the appearance of the earliest condition found, 
it would seem that there are two tiers of cells and one tier of free nuclei organised 
in the base of the archegonium. The lower tier of cells is the embryo itself. 
The next tier forms the suspensor and the tier of nuclei lies freely in the cyto- 
plasm of the archegonium immediately next to the suspensor. The latter deve- 
lops immediately and elongates quite rapidly, pushing the embryo cells deep into 
the prothallial tissues. Text-fig. 29 shows one of these early stages in longi- 
tudinal view. 

The suspensor cells, of which there are three or four to each embryo, have 
elongated to several times their width. Text-fig. 30 shows about the same con- 
dition, but the cells of the suspensor show a curvature or twist as they advance 
into the prothallium. It should be noted that there was no evidence of the plug 
at the base of the archegonium next to the suspensor as reported by Coker (1902^ 
for Podocarpm and also by the writer in Pseudotsuga (1909). Nothing in the 
nature of such a modification of the suspensor was found. The suspensor ceU 
seemed to elongate directly into the enveloping tissue. The dissolving of the pro- 



teristic appearance of the thick megaspore membrane. Feb. 6th, 
1922. 

Text-fig. 27. — This figure also represents the section through the upper half of the 
prothallus. The pollen tube, with its contents, is shown. Also the 
characteristic single group of archegonia enveloped in a common 
jacket of nourishing cells. The characteristic arrangement of the 
cells of the prothallus is shown, as also is the megaspore membrane. 
Feb. 6th, 1922. 

Text-fig. 28. — This represents a section of a mature archegonium much more 
highly magnified. Two cells of the neck are shown, as well as the 
characteristic position of the egg nucleus and the large vacuole. 
The surrounding jacket cells which appear in two layers are also 
indicated. Dec. 10th, 1922. 

(Figs. 25, 26, 27. x550; fig. 28. x800) 



* Coulter and Land, 1905. 



190 



THE LIFE-HISTORY 0¥ MICROCAOHBYS TETRAGONA;, 



thallial cells in toucli with the embryo proper was quite evident. There was all 
the appearance of enzyme action on the disintegrating cells, to make room for and 
at some time nourish the advancing embryo. I was unable to detect the presence of 
a cap cell which has been reported for Podocarpus and other genera. Text-flg. 




34 



Text-fig. 29. — A longitudinal section to show a young embryo. The suspensor 

and embryo cells are shown. Feb. 19th, 1922. 
Text-fig. 30. — A similar view of the embryo with its suspensor. Feb. 19th, 1922. 
Text^fig. 31. — A longitudinal view of an older embryo showing the elongation of 

the suspensor. The breaking down of the prothallial cells in front 

of the embryo is also shown. Feb. 19th, 1922. 
Text-fig. 32. — Two embryos are here represented. They have developed from 

different archegonia and appear to be developing simultaneously. 

Feb. 19th, 1922. 
Text-fig. 33. — This represents a slightly older condition of the embryo with the 

suspensor forcing it deeply into the prothallial tissues. Feb. 19th, 

1922. 
Text-fig. 34. — An older stage of the same with the great length of the suspensor 

shown. Feb. 27th. 1922. 

(Figs. 29-.34. x300) 



BY A. A. LAW SON. ^ ' 191 

31 represents the suspensor and the embryo further advanced. The space be- 
tween the embryo and the disintegrated or digested prothallial cells is quite evi- 
dent. I was unable to find any branching from the embryo to form embryonal 
tubes or of more than one embryo from a single archegonium. At this early 
stage, however, more than one embryo was formed in the endosperm. These, 
however, were invariably fou.nd to have originated in different arehegonia. I am 
therefore of the opinion that polyembryony does not exist other than that des- 
cribed. Text-fig. 32 shows two embryos with their own suspensors advancing to- 
gether in the endospeinn. It is quite clear that these two come from two separate 
arehegonia. It is quite probable that all the arehegonia become fertilised and 
that initial embryos are developed in them. But one of these, as a rule, advances 
at the expense of the others. Text-fig. 30 shows the embryo slightly older with 
the suspensor still straight and growing in a median line through the endosperm. 
Text-fig. 34 is still oldei', showing the great length of the suspensors. This was 
the oldest embryo observed. 

This brief account of the embryo of Microcachrys is by no means complete, 
but sufiicient has been observed to show that it differs in many interesting ways 
from that of Podocarpus. 

A very usefiil summary of our knowledge of the embryo of the Conifers has 
recently been given by Buchholz (1920). In this a comparative study of the 
Podoearpineae has been given. 

From this and other investigations (Young, 1907-1910; Robertson, 1904-1907; 
Thomson, 1909; Kildahl, 1908; Coker, 1902; Burlingame, 1908) it is quite evident 
that the Podoearpineae is not a natural class of the Taxineae and our ideas in 
regard to it require revision. The position of Microcachrys in this connection wiU 
be discussed by the avithor in another paper in the near future. 

Summary. 

1. There are three wings to the pollen grains and these are symmetrically 
arranged. 

2. The mature microspore at the time of pollination contains two prothallial 
cells, one large centrally situated body-cell, a stalk nucleus and a tube 
nucleus. 

3. In the pollen tube the body cell enlarges and, by division, gives rise to two 
small gametes. These are approximately of equal size and both are pro- 
bably functional. 

4. There is a single megaspore mother cell organised. 

5. After the reduction division four potential megaspores are produced form- 
ing an axial row. 

6. There is a very distinct and definite tapetum formed during megasporo- 
genesis. This is two or three layers of cells thick. 

7. Only one of the four megaspores germinates. The other three become 
aborted and finally absorbed. The lowest of the axial row is the functional 
spore. 

8. As the functional spore increases in size a large central vacuole is formed 
which forces the cytoplasm into a parietal position. 

9. Free nuclear division proceeds and the nuclei are distributed in the parietal 
cytoplasm. 

10. Endosperm formation takes place by the "alveoli" method. That is, primary 
prothallial cells are formed in the parietal cytoplasm which gi-ow out into 



192 THE LIFE-HISTORY OF MICROCACHRYS TETRAGONA^ 

the central vacuole in the form of long tubes. These eventually close in on 
the vacuolar space and then regular cellular tissue follows. 

11. A very distinct and thick suberised megaspore membrane is formed quite as 
thick as in the Abietineae. 

12. The archegonia, may be five or six in number, form a single group at the 
apex of the prothallus. They have a common jacket and a common arche- 
gonial chamber. There are four neck cells in one tier. 

13. Fertilisation takes place by the contents of the pollen tubes being discharged 
into the archegonial chamber. 

14. The embryo is organised from three tiers of nuclei in the base of the arche- 
gonia. The middle tier forms the suspensors which carry the embryo into 
the endosperm. 

In conclusion I desire to express my thanks to my Laboratory assistant, Mr. 
0. D. Evans, for help in the collection and preparation of material. 

I also desire to express my indebtedness to the University of Sydney for a 
grant of money from the McCaughey Research Fund, which helped to defray the 
expenses in the collection of specimens in Tasmania. 



Literature Cited. 

Arnoldi^ W., 1900. — Beitrage zur Morphologie der Gymnospermen, III. Em- 

bryogenie von Cephalotaxus Foriunei. Flora, 87, 46-63. 
BuRLiNGAMEj L. L., 1908. — The staminate cone and male gametophyte of Podo- 

carpibs. Bot. Gazette, 46, 161-178. 
COKER, W. C, 1900.- — Observations on the gametophyte and embryo of Taxodium 

distichum. Joh/ns Hopkins Univ. Circ, 19, 45-46. 
, 1902. — Notes on the gametophytes and embryo of Podocarpus. 

Bot. Gazette, 33, 89-107, Pis. 5-7. 
— , 1904. — On the spores of certain Coniferae. Science, N.S., 19, 

124. 
, 1907. — Fertilisation and Embryogeny in C eplialotaxus Fortunei. 

Bot. Gazette, 43, 1-10. 
Coulter^ John M., and Land, W. J. G., 1905. — Gametophytes and Embryo of 

Torreya taxifolia. Bot. Gazette, 39, 161-178. 
Hooker, J., 1845. — Lond. Jour. Bot., IV., 150. 
Jager, L., 1899. — Beitrage zur Kenntniss der Endospermbildung und zur Em- 

bryologie von Taxus haceata. Flora, 86, 241-288. 
KiLDAHii, N. Johanna, 1908. — The Morphology of Phyllocladus alpina. Bot. 

Gazette, 46, 339-348, Pis. 20-22. 

, 1908a.— Affinities of Phyllocladus. Bot. Gazette, 46, 464-465. 

Lawson, a. a., 1904. — The gametophyte, archegonia, fertilisation and embryo of 

Sequoia sempervirens. Ann. Bot., 18, 1-28. 
— , 1904a. — ^The Gametopliyte, fertilisation and embryo of Crypto- 

meria japonica. Ann. Bot., 18, 417-444. 
, 1907. — The Gametophytes, fertilisation and embryo of Cephalo- 
taxus drupacea. Ann. Bot., 21, 1-23, Pis. 1-4. 
1907a. — The gametophytes and embryo of the Cupressineae, with 

.spocial reference to Lihooedrus decurrens. Ann. Bot., 21, 281- 

301, Pis. 24-26. 



BY A. A. LAWSON. 193 

, 1909. — The Gametophytes and embryo of Pseudotsuga Douglasii. 

Ann. Bot., 23, 163-180, Pis. 12-14. 
, 1910. — The gametophytes and embryo of Sciadopitys verticillata. 

Ann. Bot., 24, 403-421, Pis. 29-31. 
Lopri6re_, G., 1905. — Ueber die Vielkornigkeit der PoUenkorner von Araucaria 

Bidwillii. Vorlaufige Mitteilung. Ber. Deutsch. Bot. Gesell., 

33, 335-346, PL 15. 
MiTAKE, K., 1903. — On the development of the sexual organs and fertilisation in 

Picea excelsa. Ann. Bot., 17, 351-372, Pis. 16, 17. 
, 1908. — The development of the gametophytes and embryogeny 

of Cunninghamia (preliminary note). Bot. Mag. Tokyo, 22, 

45-50, fig. 14. 
, 1910.— The development of the gametophytes and embryogeny in 

Cunninghamia sinensis. Beih. Bot. Centralbl., 27, 1-25, Pis. 1- 

5, figs. 2. 

NOREN, C. 0., 1904." — • Ueber Befruehtung bei Junipervs communis. Vorlaufige 

Mitteilung. ArMv Bot. Svensk. Vetensk.-Akad., 3, 1904. 
• , 1907. — Zur Entwiekelungsgeschichte des Juniperus communis. 

Uppsala Universitets Arsskrift, 1907, pp. 64, Pis. 4. 
, 1908. — Zur kenntnis der Entwicklung von Saxegothaea conspicua. 

Lindl. Svensk. Bot. Tidsskr., 2, 101-122, Pis. 7-9. 
Robertson, Agnes, 1904. — Spore formation in Torreya calif ornica. New Phytolo- 

gist, 3, 133-148, Pis. 3, 4. 
, 1907. — The Taxoideae; A phylogenetic study. Neiv Phytologist, 

6, 92-102, PI. 1. 

Saxton, W. T., 1909. — Preliminary account of the development of the ovule, 

gametophytes and embryo of Widdringtonia cupressoides Endl. 

Bot. Gazette, 48, 161-178, PI. 11, ■ 
, 1910. — Contributions to the life-history of Widdringtonia cupres- 
soides. Bot. Gazette, 50, 30-48, Pis. 1-3. 
, 1910a.— Notes on the anatomy of Widdringtonia and Gallitris. 

S. African Jour. Sci., 7, 282-286, fig. 11. 
Thomson, R. B., 1905. — Preliminary note on the Arauearineae. Science, N.S., 

22,88. 
, 1908. — Note on the pollen of Microcachrys. Bot. Gazette, 46, 

465, 466, 1908. 
, 1909. — On the pollen of Microcachrys tetragona. Bot. Gazette, 

47, 26-29, Pis. 1, 2. 
, 1909a. — The megasporophyll of Saxegothaea and Microcachrys. 

Bot. Gazette, 47, 345-354, Pis. 21-25. 
Young, Mary S., 1907. — The male gametophyte of Dacrydium. Bot. Gazette, 44, 

189-196, PI. 19. 
, 1910. — The morphology of the Podocarpineae. Bot. Gazette, 50, 

81-100, Pis. 4-6. 

EXPLANATION OF PLATES XV.— XVI. 

Plate XV. 

Microcachrys tetragona, showing general habit of plant with microspore cones (nat. 
size). 

Plate xvi. 

Microcachrys tetragona. Branches showing megaspore cones. (x2.) 



194 



STUDIES IN SYMBIOSIS. 

IV. The Root-nodules oj? Casuarina Cunninghamiana and their 
Physiological Signieicance. 

By J. McLucKiE^ M.A., D.Sc, Lecturer in Plant Physiology, University of Sydney. 

( Sixteen Text-figures. ) 

[Read 30th May, 1923.] 

Introduction. 

Since the discovery by Hellriegel and Wilfarth in 1888, that the root-nodules 
of the LegTiminosae are able to "fix" or assimilate atmospheric-nitrogen, much re- 
seai'ch has been conducted, not only upon the various genera of the Leguminosae, 
but also ujDon other root-nodule possessing types, such as the Cycadales, Alnus, 
Eleagnus, Myrica Gale and Podocarpus. Early investigators of the subject did 
not find bacteria in the nodules, and believed that the fungal hyphae, which com- 
monly occur in older nodules, constitute a mycorhiza. In Podocarpus, Nobbe and 
Hiltner (1899) proved indirectly by experiments that the root-nodules were 
active nitrogen-fixing structures, but they and others (e.g., Shibata, 1902) be- 
lieved this property to be due to what appeared to be fungal hyphae in the 
cortical cells of the nodules. It was not known that the active nitrogen-fixing 
organisms are bacteria which migrate from cell to cell in zoogloea threads bear- 
ing certain resemblance to fungal hyphae. 

In 1896 Hiltner demonstrated the presence of bacteria in the root-nodules of 
Alntis and Eleagnus, and in a series of experiments proved (1) that plants de- 
void of root-nodules could not thrive in a soil poor in nitrogen, (2) that such 
starved plants thrived when the soil was inoculated with the organisms from the 
nodules, and (3) that the nodule-formation increased with the poverty of the 
soil in available nitrogen. 

In 1904, Life investigated the root-nodules of Cycas and discovered bacteria 
within them. Bottomley (1907) isolated these bacteria, and proved them cap- 
able of nitrogeji-fixation. Bottomley (1912) studied the development and struc- 
ture of the root-nodules of Myrica Gale, and isolated Pseudomonas radicicola 
from them; Spratt (1912) perfonned similar researches upon Alnus and Eleag- 
nus, and claims that Pseudomonas radicicola occurs in these nodules also, and that 
it is polymorphic. McLuckie (1922) investigated the tubercles of Macrozamia 
spiralis and isolated bacteria from them and proved that they wei'e nitrogen-fixing 
fonns. In the present communication, the author gives the result of his investi- 
gations of the root-nodnles of Casuarina Cnnnin ghamiana , a form which has not 
hitherto been descril)f'd in detnil. Darnell-Smith in a brief note mentions 



BY J. MCLUCKIE. 



195 



their occurrence and that they are inhabited by bacteria, but no detailed account 
of the nodules is given. 

The Boot-nodules. 

Casuarina Cunninghamiana is a species which commonly occurs along the 
banks of creeks. The material for this investigation was obtained along the banks 
of the Nepean River near Penrith, N.S.W. The gTeat majority of the roots of 
this plant possess nodules of varying sizes, some being quite small and branched 
once or twice, others being profusely branched, 50-60 mm. in diameter, and larger 
than any other root-nodules yet recorded. The youngest nodules occur as small 
lateral growths upon the youngest roots, and are from 3 mm. to 5 mm. long and 




Text-fig. la. — Roots of Casuarina Cunninghamiana (M'iq.) bearing nodules of 

various sizes, (x 4.) 
Text-fig. lb. — Single large coralloid nodule showing the rootlets emerging from the 

cluster, (x \.) 
Text-fig. Ic. — Portion of a coralloid mass showing method of branching, (x 2.) 

1-2 mm. in diameter. They bear a general resemblance to the young nodules of 
Maerozamia or Cycas at this stage. Branching beginsi early in the primary no- 



196 STUDIES I2ST SYMBIOSIS, iv., 

dules, and from the ineristematic cells at the apex, three or four groups are 
formed, which become the apical meristems of the branches. These again branch 
after a short growth phase, and in a comparatively short time a remarkably large 
''cluster" nodule is formed (Text-fig-s. la, lb). The nodules are perennial and 
are covered externally with several thin layers of cork. In nature many of the 
nodules are exposed or lie very close to the surface of the soil, as in Macrozamia 
and Cycas. The general appearance of the mature eoralloid nodules is similar to 
that of the root-nodule of Myrica Gale described by Bottomley. As seen in Text- 
figs. 1«, 16, numerous short thin rootlets project from the nodule, but die ofl: ter- 
minally. The tips of the nodular branches are rather whitish, while the rest of 
the structure is protected by cork. Text-fig. Ic, shows a small portion of a large 
eoralloid nodule and incidentlj^ the branching which is typical of this form. 

For anatomical work the material was fixed either in 1% chromaeetic acid 
or in 10% acetic alcohol. The stains found most useful were Gentian Violet and 
Safranin; Gentian Violet, Lugol's Iodine and Safranin, and Gram's stain. Median 
longitudinal sections of the nodules are shown in Text-figs. 2, 3, 4, and 5 ; a central 
stele is plainly demarcated, surrounded by an endodennis containing a tannin de- 
posit. A clearty defined meristematic zone is seen at the apex, and in Text-fig. 2, 
it has bifurcated. In Text-figs. 3, 4, and 5 different stages in the branching of 
the nodnle may be seen. The meristematic zone of cells generally divides into 
four fairly distinct groups: namely, a central mass which continues the growth 
of the axis of the mother root, and three, more or less laterally developing masses 
which form the branches of the primary nodule. In a longitudinal median sec- 
tion, as a rule, only three of these groups and the stelar structure passing into 
them are seen. The apical part of each nodular branch becomes considerably 
swollen. Beyond the apical meristem groups, several layers of small cells are 
formed, which, however, cannot be recognised as a definite root-cap, and which 
merge imperceptibly into the cortical tissues behind the apex. In young branches, 
the extreme tip of the nodule is not covered with cork, but the apex of all the- 
older nodules is so protected, and the surface becomes quite hard and brittle. 
The apical zone of meristematic cells, or their immediate derivatives, are free from 
bacterial infection. Behind the meristematic zone three distinct zones are de- 
fined, namely (a) the cork (c), (b) the cortex {cor), and (c) the stele. In the 
cortex, groups of bacterial cells occur; these infected eeUs are large, thin-waUed, 
and densely cytoplasmic. In sections of living nodules the bacterial ceUs are 
differentiated by their yellowish contents and in fixed material by the staining 
of the bacteria. In the cortical cells of the nodule three definite regions may be 
observed, namely, (a) the zone of infection {z. inf.), (b) the mature zone of 
bacterial masses {m.z.), (c) the degenerating zone where bacterial digestion oc- 
curs (d.z.). Text-figures 6, 7, 8, 9, show various stages of infection of the 
younger cortical cells. A few or numbers of infection threads may be seen pass- 
ing from cell to cell, and coiling in the vicinity of the nucleus. These threads 
contain rod-shaped bacteria, and have a rather slimy consistency. They are of 
irregular diameter. In some cells, especially those just infected, these threads are 
very thin, but in slightly older cells, the bacterial threads are thicker and more 
numerous. The infection of a cell by the bacteria leads to a considerable en- 
largement of the cell so that, apart from the difference in their contents, they are 
readily distinguished from uninfected cells by their size. A very important 
feature of the bacterial zone is the large intercellular spaces providing for gaseous 
circulation. These spaces communicate with the atmosphere through the irregu- 
larities and cracks in the superficial lining of cork cells, or through the epidermal 



BY J. MCLUCKIE. 



197 



cells near tlie apex. The bacterial ceils are divided into groups by longitudinally- 
arranged fibres, or thiekeaaed parenchymatous cells, which frequently take an 
irreg^ilar course in the midst of the thin-walled tissue. These points are clearly 




Text-figs. 2, 3, 4, 5. — Longitudinal sections of apex of nodule showing the branch- 
ing, the corky covering (c), the cortex (oor.) inhabited with bacteria, 
the stele. The infected zone (inf. z.) is behind the apex, the mature 
zone (m.z.) further back, and the zone where bacterial degeneration 
occurs (d.z.) is the older part of the nodule, (x 33.) 



198 



STUDIES IN SYMBIOSIS, IV., 



brought out in the Text-figures 2, 3, 4, 5, and 10. In old bacterial cells, the 
threads can seldom be distinguished, and the bacteria appear to be more or less 
uniformly distributed throughout the eell-cavity (Text-figs. 11 and 12). 

Near the base, i.e., in the older part of each nodular branch, the bacterial 
cells undergo a change. Many of the bacteria either leave these cells, or undergo 
degeneration into baeteroids. They swell up, lose their contents, and disinte- 
grate. This degeneration of active bacteria into inactive baeteroids is apparently 
a natural event, as in every culture of the bacteria upon nutritive turnip or 
glucose agar, or in liquid media as demonstrated later, many bacteria become 




Text-figs. 6, 7, 8, 9. — Cells of infection zone (inf. z. in previous figures) showing 
the thin gelatinous threads containing the bacteria and spreading from 
cell to cell, (x 600.) 



quite inactive and enlarged although the culture conditions, such as food supply 
and temperature, were constant. Throughout the cortical region of the nodules 
there is abundance of food. 

In the nodules of Podocarpus spinulosa, I have observed fungal hyphae in 
the older cortical cells, previously infected by bacteria,, but in Casuarina I hav^ 
not seen fungal hyphae in the tis.sues. 

In infected cells, the nucleus does not appear to be stimulated to mitotic 
activity, as is frequently the case in fungal-infected cells, but the chromatin 
aggregates from the finely reticulated condition to a coarse granular state. Fre- 
quently two or three nucleoli appear in the nucleus, and in a few cells two nuclei 
were observed. In many of the old cells containing the baeteroids, one or more, 



BY J. MCLUCKIE. 



199 



globular yellowish masses of an oily nature were observed but no starch. Pro- 
bably these masses are the residues of digested baeteroids. 

Text-fig. 13 shows a transverse section of a nodule, through the bacterial 
zone; the cork, cortex, and stelar zones are readily distingaiished. The stele is 
tetrareh. Surrounding the stele is the clearly defined endodermis. This layer 
and others surrounding it contain a quantity of tannin. 




Text-fig. 10.— A small portion of the cortex from a longitudinal section of nodule 

showing the bacterial cells (b.c), the starch cells (s.c.) and the fibrous 

cells (f.c). (x 167.) 
Text-figs. 11, 12. — Old cells of the cortex showing dense masses of bacteria; in 

such a stage the threads can be seen only with difficulty, s.c. starch 

cells, (x 600.) 

The cortex is composed of enlarged parenchyma cells containing bacteria, 
prosenchymatous or fibrous cells with thickened walls, and thin-walled cells con- 
taining food such as starch. In cells showing the infection threads considerable 
numbers of smaU starch grains are observed, but these later disappear, and are 
not present in the cells containing baeteroids. Numerous cells containing a 
brownish red tannin deposit occur throughout the cortex. 



Nodule a modified lateral root. 

The coralloid nodule is an aggi-egation of short branches derived from a 
primary modified lateral root whose normal development is inhibited by the bac- 
terial infection. 

The bacteria gain access to the cortical tissues of the main root through the 
root-hairs. The young lateral root is frequently infected soon after its inception, 
b.y the migration of bacteria from the cortical cells into its periblem cells. This 
infection retards the normal elongation of the root, but its diameter increases at 
first by the enlargement of the infected cells and the normal growth of the stelar 



200 



STUDIES IN SYMBIOSIS^ iv., 



tissue, but later by the meristematic activity of a phellogen which superposes 
corky layers upon the older parts of the root and increases the cortical tissue 
within. The nodule, therefore, starts as a modified lateral root. The presence 
of bacteria just behind the meristematic zone appears to stimulate vigorous and 
frequent branching such as we do not find in the normal root branching. There 
is no root-cap upon the nodule, but some of the apical tissue of the original 
primary root-nodule is continuous over the secondary branches. This is seen in 
Text-fig. 5. 




Text-fig. 13. — TjS. of nodule showing stele, cortex and cork, and the distribution of 
the bacterial cells. (x50.) 

The central group of meristematic cells grows forward into a thin rootlet 
which soon dies off at the free end and becomes thickly covered with cork. 

Around the primary nodule, three lateral secondary nodules are developed, 
while the axis of the original nodule continues to grow forward as a thin fibrous 
root. Each secondary nodule enlarges and subsequently forms two or three ter- 
tiary nodules, leaving the main axis, as before, to grow out into a thin rootlet. 
In this manner a large coralloid mass is soon formed. 



The nodule bacteria. 

Young nodules were sterilised carefully by immersing several limes for a 
minute in a sterilising fluid employed by Harrison and Barlow when investigating 
the nodules of the Leguminosae, and washing after each immersion, in distilled 



BY J. MCLUCKIE. 



201 



"water. The composition of the fluid employed is as follows : — Distilled water 500 
e.C; Mercuric chloride 1 gram, Hydrochloric acid 2.5 ec. After careful sterilisa- 
tion and washing, one of the nodules was crushed upon a slide, the smear was 
dned and stained in Carbol Fuehsin, and in Gentian Violet, Lugol's Iodine and 
Safranin. The bacteria proved to be small rod-shaped forms, bearing a very 
close resemblance to the organisms in Podocarpus spinulosa. 

When stained with Kiskalt's amyl gram preparation, the bacteria were readily 
decolorised with ethyl alcohol, but the gentian violet is retained with the amyl 
alcohol. From this fact, and the general form and life-history of the bacteria I 
am inclined to regard the organism of Casuarina as a species of Pseudomonas. 



^^Q & ff 



^ . ^ 




14 




Text-fig. 14. — The active rod-shaped bacteria, (x 1400.) 

Text-fig. 15.— A group of cortical cells from the degenerating zone (d.z. of Figs. 

2, 3, 4, 5) showing the bacteroids (b) and the oily masses (o.m.). (x 

1400.) 



Pure cultures of the organism were made by transferring some of the ex- 
pressed contents of sterilised nodules by means of a sterile platinum needle to 
sterilised turnip agar or glucose agar in test tubes and in petrie dishes. The 
tubes and dishes were incubated at a temperature of 26. 5° C and in the course 
of 24 to 48 hours small vigorously growing colonies of the bacteria were pro- 



202 



STUDIES IN SYMBIOSIS^ IV., 



duced. The colonies were generally roundish, raised, exceedingly viscous and 
shining on the surface (Text-fig. 16). They enlarged very rapidly. By means 
of a platinum needle some of the bacteria were transferred to a clear sterilised 
glass slide, and on the smear being dried and stained the typical rod-shaped bac- 
teria of the living nodules were observed. Text-fig. 14 shows the typical character 
of the bacteria of Casuarina Gunninghamiana, and it will readily be observed 
that they are remarkably similar to those of Podocarpus spinulosa, and of Alnus 
(as figured by Spratt, 1912). 

Parts of old nodules were similarly sterilised and their contents expressed 
upon a clean sterilised slide and stained. In this case, very few of the typical 




Text-fig. 16.— Photographs of bacterial cultures from Casuarina nodules. 

short rods were found, but numerous oval, enlarged, granulated individuals were 
obser^'ed. Many of these contain granular bodies, whUe others are empty. These 
structures are probably bacteroids derived by degeneration from the active rod 
forms. Text-fig. 15 shows a few cells of the older part of the nodule containing 
the typical bacteroids of this form. 



Nitrogen -fixation by the bacteria. 

The culture solution for the purpose of this estimation had the following 
constitution : — Saccharose 10 grams. Acid potassium phosphate 5 grams. Mag- 
nesium sulphate 0.5 grams, Distilled water 1000 c.c, neutralised with sodium 
hydrate. Each of five 500 c.c. flasks, Al, A2, . . , containing 100 c.c. of this 



BY J. MCLUCKIE. 203 

solution was autoclaved for 20 minutes at a temperature of 140° C, cooled and in- 
oculated with the bacteria from a young culture upon turnip agar, by means of 
a platinum needle. Two flasks Bl and B2, each containing 100 c.c. of the 
nutrient solution were inoculated <ind autoclaved in the same way. These con- 
stitute controls. 

All flasks were then incubated at a temperature of 26. 5° C. for 14 days, 
when a Kjeldahl estimation of the nitrogen content of the flasks was made. The 
following results were obtained. 

In the flasks Bl and B2, in which the organisms were killed by autoelaving, 
only a slight trace of nitrogen was found to be present. 

In flasks Al, A2, A3, A4, A5, the nitrogen content was respectively 5.45, 
5.1, 5.82, 5.32, and 5.65 mg. The average per flask is 5.468 mg. for 14 days, 
or average fl:sation of nitrogen amounting to 0.3906 mg. per day. As the flasks 
at first contained no combined nitrogen, excepting that introduced during in- 
oculation with the bacteria, which proved, in the controls, to be a mere trace, the 
Dacteria introduced to the culture solution must have ''fixed" or assimilated the 
free atmospheric nitrogen of the flask. 

In the course of these experiments, it was observed that nitrogen-fixation was 
more active in large flasks in which the nutritive fluid could be spread out in a 
thin layer. When flasks of 300 c.c. capacity were employed, the nitrogen fixation 
per day was only 0.32 mg. The temperature found to be most suitable was 
26.5°C. 

It therefore appears that the nodules of Casuarina Cunninghamiana contain 
bacteria v/hich possess the power of fixing free atmospheric .nitrogen, and, from 
the analogy of the Leguminosae, that the plant benefits by the presence of the 
root-nodules. 

Nutrition of the bacteria. 

When the bacteria are grown in a nutritive culture solution they multiply 
rapidly, imparting a cloudy appearance to the solution. During metabolism they 
assimilate free nitrogen from the flask and convert it into an organic state which 
is availed of by the host-plant as in the Leguminosae. But the bacteria require 
carbohydrate during this process. Experiments show that they develop just as 
actively in the presence of glucose or maltose, or saccharose, while their activity' 
is inhibited if saccharose or glucose or maltose is omitted from the solution. They 
cannot obtain their carbon from the CO2 of the flask. In the bacterial cells of 
the root-nodules, the bacteria no doubt obtain supplies of carbohydrates manu- 
factured by the host. A considerable amount of starch may be observed in cells 
which ai'e free of bacteria or in the early stages of infection, but in mature bac- 
terial cells, when the bacteria come to occupy the whole cavity of the cell, no trace 
of starch can be found. It is probable, however, that it is transformed to sugar 
(maltose) and as such is absorbed by the bacteria. 

Although the nodules of Casuarina are perennial and protected with cork, 
sooner or later portions of the coralloid mass begin to decay and the bacteria or 
baeteroids which they contain are given up to the soil. By this means the soil 
becomes enriched with combined nitrogen after the manner of the Leguminosae. 

Summary. 

1. The nodules on the root of Casuarina Cunninghamiana are modified lateral 

roots whose normal development is modified by bacterial infection. /"O' "os 






204 STUDIES IN SYMBIOSIS^ iv., 

2. The nodules branch profusely, three or four branches being formed from 
the apical meristem of the primary nodules. By the similar branching of 
these secondary nodules and their derivatives, a large "coralloid" mass of 
nodules is formed which may attain very considerable dimensions. 

3. From the coralloid mass, slender fibrous rootlets radiate; these die off 
terminally. 

4. In each nodule there is a covering of cork, a comparatively wide cortex com- 
posed of large parenchymatous cells containing bacteria, and longitudinal 
rows of fibrous cells, and a central stele surrounded by an endodermis. 

5. The meristematic tissue does not contain bacteria. 

6. In each nodule there is (1) a region near the apex, behind the meristem, 
containing infection threads of the bacteria, (2) the extensive zone of bac- 
terial cells divided by the fibrous cells into groups, and (3) the basal zone 
which contains the older bacterial cells in which the bacteria are inactive and 
converted to ''bacteroids." 

7. The nodule has no root-cap, but owing to the peculiar method of growth 
some of the tissue of the primary nodule covers the apex of its derivative 
branches. 

8. The nodules are perennial and, with the exception of a small apical portion 
on the youngest, are <3ompletely covered with cork. 

9. The bacteria were isolated in pure nutritive agar cultures and examination; 
revealed that they were small rod-shaped forms somewhat closely resembling 
Pseudomonas radicicola which occurs in leguminous nodules. 

10. The rod form becomes transformed into an oval baeteroid, which may 
undergo digestion by the cell-cytoplasm. 

11. The bacteria were grown in liquid cultures free of combined nitrogen, and 
containing carbohydrate. Kjeldahl estimates of the nitrogen content of the 
flask demonstrated that an average of 5.468 mg. of N. were "fixed" in 14 
days per flask, or . 3906 mg. per day. 

12. Saccharose, maltose and glucose proved of practically equal nutritive value 
to the bacteria, which are unable to use the CO2 of the air in the flask as a 
source of carbon. 

My sincere thanks are due to Professor Lawson for his kindly advice and 
criticism, and to Mr. 0. D. Evans for his trouble in fmding some of the material. 
Botany School, Sydney, January 25th, 1923. 

Literature. 

BOTTOMLET, W. B., 1907. — The structure of Eoot Tubercles in Leguminous and 
other plants. Beport British Association, 693. 

, 1912. — The Root Nodules of Myrica Gale. Ann. Bat., xxvi., 111- 

117. 

Darnell-Smith, G. P. — -Investigations on the Soils of N.S.W. in relation to Agri- 
culture, with special reference to the role of Nitrogen and 
Phosphorus. Report of the Bureau of Microbiology, Depart- 
ment of Agriculture, N.S.W. 

Harrison and Barlow, 19,06. — The Nodule-Organism of the Leguminoseae. Proc. 
Roy. Soc. Canada, 1906. 

Hellriegel and Wilfarth, 1886.— Unters. ul)or die Stickstoffnahrung der Grami- 
neen u. Leguminosen. Beil. Zeit. d. Vereins fiir die Ruhen- 
.zuckerindustrie, Berlin . 



Prog. Linn. Soc. N.S.W., 1923. 



Plate i. 




Approximately four-fifths natural size. 



Pkoc. Lixx. Soo. N.S.W., 1923. 



Platk n. 




Proc. Liyn. Soc. N.SAY., 1923. 



P[,ATE III. 




\^'.F.B. del. 



Loraiiilius Qitcenslandicus, u.sp 



pROC. LixN. Soo. N.S.AV., 1923. 



Plate iv. 




f !0 



A. Lorantlins Cvcneiis-Siiius, n.sjj. B. L. Mackavensis, n.sp. 
W.F.B. del. 



Proc. LnvN. Soc. N.S.W., 1923. 



Plate v. 




W.F.B. del. 



LoriDithus iiiiraciilosus Miq. 



Proc. Link. Soc. N.S.W., 1923. 



Plate vi. 



^ Ufl /) 




LoraiitJiui miracu/osiis Miq. var. Melalencae (Tate), n.conili 
W.F.B. del. 



Proc. Lin^-. Soc. N.S.W.. 1923. 



Plate vrr. 




1-7. LorantJuis iniraciilosiis Miq. var. Boonnani. 
var. piibigera , n.var. 
^^■.F.B. del. 



8, 9. L. iniracidosus 



PROC. LixN. Soc. N.S.W.. 1923. 



1'lATE VIII. 




Loraiifhns Gaudicliaiidi DC. 



W.F.B. del. 



ruoc. Lixx. Soo. N.S.W., 1923. 



Plate ix. 




W.F.B. del. 



Loranthiis Preissii Miq. 



Proc. LiXiS\ Soc. N.S.W., 1923. 



Plate x. 




W.F.B. del. 



Loi-a)iilius Caiiibagei, n-sji. 



pROc. LixN. Soc. N.8.W., 1923. 



PIjATR XI. 




W.F.B. del. 



Loraiil/iiis~Jinop/iyl/i(s Fenzl. 



Proc. Linn. Soc. N.S.W., 1923. 



Plate xti. 




W.F.B. del. 



Loranthns conspiciiiis Bnil. 



Pkoc. Likn. Soc. N.S.W., 1923. 



PlATK XIII. 




A. Loranthus Beic/iei, n.sp. B. L. Betchei, var. duhia, u.var. 

C. L. fletc/iei, var. ionieiitil/a , u.var. 



W.F.B. del. 



Proc. Linn. Soc. N.S.W.. 1923. 



Plate xiv. 




W.F.B. del. 



Loraiithus obliqtia, n. sp. 



o 



CL 




►J 
(In 



6 
o 

:z; 

^3 




o 

o 

O 
O 



- ^ I CO W 



^ .s 



■I- 



■gns 






i 






s^ a^ 






H 01 rH cs te 



ccl 



^1 a. 



i^^.gsl.a 



O 



=fH 



=H 



o 1^ o 

lO t-* O CO 
tH ^ O <^ 



0) 



o » o 

0) <S O C 
o3 ^ « 



3 CO r") 
> So S 






03 1^ 



M .la 



03 



X5 43 



cc C/3^ CS 
03 <:d ~ 

03 5d e 

l-l. ^ 

T « e f^ 

^ S o '§ 



Co ?; fv 






o cs 









«C 



IK 



^ 



^ CO 



^ £3 m 03 
-H Si 03 r> 

o 2f S 



03 r^a 

03 

a 
Q 



>^C/2 



03 -^ 



-^l-^i 



PQ 



03 N 
ra (3; Oj 



^ ft 



."rt ° 






CC El 

IS 



o 



tt-H 



;^ 



O! O '^ 

03 o 00 



CO 



cc 



03 ." 

o ^^ 



o '^ 



ft 



o3 o 

03 o 
gco 

o e~ 

U 03 
03 ro 

^ s 

O Oj 



n3 



Oj 



a^ 



ShCq 



R^ « 



Ip^' 



a. 









CO rO 



f*5 

o P 

«^ 

03 P 
c3 ® 

03 -r^ 

log 



g S 



a.=>8 



m 



ft 



S!=3 (=1 



^ r5 f-l ft 

O O r> 



03 03 j2 






ft 
ft 



-^ -2 



2 CO 



•^S (U 



O o '-' 



I? 



+J 03 



m 



^ 




ce 




-»j 




02 




■Xi 




Pl 


/^•^^ 


o3 






« 


03 


•C 


"ft 


-(-3 



e 











M -(J 




-►J 


PI 
o 








+? 




(Middle ; 
ferous) 
r beds w 




03 
03 


1 
1 ^' 


^ 




£ 


> 

03 03 


'a 


?eds 
irboni 
Uppe 
main? 




remai 
Lowe: 
basic 


o 


^O^ 




,-^ 


^_^ 


u a 




-Q 


u 


a 


a ^ 




^*^ 


^-^ 


P 


CQ 









2^ ^ 



^S 



P.-3 



p ^ 



* V 



BY J. MCLUCKIE. 205 

HiLTNER 3896. — Uber die Bedeutung der Wurzelknollehen vou Alnus. Landw. 

Versuchsstat., Bd. 47. 
LiFE_, A. C, 1904. — The tubercle-like roots of Cycas revoluta. Bot. Gaz., 31, 265- 

271. 
McLucKiE, J., 1922. — The Apogeotropic Roots of Macrozamia spiralis, and their 

Physiological Signifieaneo. Proc. Linn. Soc. N.S.W., 47, 319- 

328. 
, 1923. — A contribution to the Morphology and Physiology of the 

Root-Nodules of Podocarpus spinulosa and P. elata. Proc. Linn. 

Soc. N.S.W., 48, 82-93. 
NOBBE and Hiltner, 1899. — Die endotrophe Mykorrhiza von Podocarpiis und 

ihre physiologische Bedeutung. Landw. Versuchsstatt., Lief. i. 
Shibata, K., 1902. — Cytologisehe Studien uber die Endotrophen Mykorrhyza. 

Jalirh. f. Wiss. Bot., 37, 643-684. 
Spratt, 1912. — The formation and Physiological Significance of the Root Nodules 

in the Podocarpineae. Ann. Bot., 26, 801-814. 
— , 1912a. — The Morphology of the Root Tubercles of Almis and 

Eleagnm. Ann. Bot., 1912, 120-128. 




206 



NOTES ON TRESHWATER ALGAE.* 

By the late G. I. Playfair. 

(Thirty Text-figures.) 

[Read 27th June, 1923.] 

i. — Chlamyclomonas globosa Snow, and allied forms. 

It was during the summer of 1918 that I obtained a pure culture of this 
species under rather curious and interesting circumstances. A small enamelled- 
iron pie-dish, with scraps of fish adhering to its sides, had got adrift on an 
almost horizontal corrugated iron roof situated in the very heart of the business 
portion of Lismore (N.S.W.), where it was exposed to the full blaze of the sun 
throughout the gTeater part of the day. The month of March (summer) opening 
with alternate heavy rain and hot sunshine, on the 3rd the dish was noted to be 
full of rainwater, on the 10th the water showed a yellow tint which next day had 
deepened to pale yellow-green. Besides, there was a certain amount of yellow- 
gi-een sediment at the bottom of the dish, and scum on the surface of the water. 

An examination made on the 11th with a 1-12 homog. imm. objective re- 
vealed a culture of Chlamydomonas glohosa Snow, easily recognized by its minute 
size, spherical shape and general delicacy of appearance. The cells were 
commonly of diam. 4-8yu,, some motile, some non-motile, very transparent, the 
chloroplast very delicate and homogeneous, a few small amylaceous granules 
round the central core of the cell. A few cells were larger, diam. 10-12^. 
Pyrenoid distinct; also stigma midway between the poles; cell-membrane very 
thin, invisible. 

Here and there spherical mother-cells (diam. c. 25/a), with 4 autospores of 
diam. 7fj, arranged tetrahedrically, revealed the method by which the rapid gro\vth 
of the culture had been attained. 

Chlamydomonas glohosa was first noted in the plankton of Lake Erie and 
has generally been considered a distinctively plankton form. However I find it 
quite common after rain in all sorts of places, e.g., roadside puddles, small pools, 
duck-ponds, swamps, and even enmeshed in filamentous tufts on grass lands or 
in fungoid growths on dead sticks in standing water. It is not any particular 
habitat that conduces to the development of plankton forms but an abundant 

f*The manuscript of this paper was completed by the author just before his 
death in October, 1922. The drawings, however, were not complete, only a number 
of rough sketches having been prepared. The figures illustrating ths paper have 
been drawn from those rough sketches.— Ed.] 



BY THE LATK G. I. PLAYFAI«. 207 

supply of water, either Iroiii the natural environment, or from rain, or from 
surface drainage. 

Tlie occurrence of a culture in quantity in a domestic cooking utensil which 
had never previously held even garden soil and which was far removed from 
any source of infection, shows how wind-blown the resting cells of the fresh- 
water algae may be. 

Var. MACULATA. n.var. 

Forma membrana tenuissima; chloroplastidibus maculatis vel scrobiculis 
notatis. Cetera ut in forma typica. Specimina parva granulis (ut videtur) 
aspera. Diam. 10-12, vel 13-16 fj.. 

Intermingled with the type were cells showing very distinctly a spotted or 
pitted chloroplast. The cell-wall being very delicate, this caused the smaller 
specimens to appear granulate. 

Chlamydomonas ovata Dangeard. 

Mem. sur les Chlamyd., p. 147, f. 17; cf. Wille, Gatt. Chlamydomonas in 
Algol. Notizen, ix.-xiv., 1903, p. 133, T. iv., f. 3. 

Yar. PULCHRA^ n.var. 

Cellulae lentieulares, utroque polo subacutae; lateribus aequaliter arcuatis. 
Long. 18, lat. 9|U,. 

A still later examination of the culture revealed no new development; but 
brought to light this beautiful and interesting form which seems to be too near 
Dangeard's Chi. ovata to be separated from it. 

Cells broadly lenticular in face view, with subacute ends and evenly arched 
sides, stigma not noted. From the shape of the mother-cell it would appear that 
this is a form produced by longitudinal self-division of a spherical cell into 4 
parts. In Dangeard's original figure (reproduced by Wille, op. cit.) the chloro- 
plast seems to be disarranged and to have shifted round to the side of the cell, 
probably with a view to a transverse division. There would thus be produced 
2 spherical cells on a smaller scale. Self-division in two directions is common 
enough in the freshwater algae generally. 

No dimensions are given for CM. ovata, but the figures woi'k out at : — Long. 
19, lat. 8-8J/i. 

Chlamydomonas Snowii Printz. 

Skr. Vidensk-Selsk. i. Kristiania; Mat.-Naturv. Kl., 1913 (1914), No. 6, p. 18. 

Along with Chi. glohosa, Miss Snow found and described (Bull. U.S. Fish. 
Comm., 1903, 22, p. 388, PI. i.) another species, viz.: Chi. comrmmis Snow. That 
name being preoccupied, the form has been renamed as above. The latter, how- 
ever, is really an oval or oblong variation of CM. glohosa. I find them always 
together. In the above natural culture this form was present in quantity; 
dimensions long. 7-10, lat. 5-7/x, stigma a little above the centre and the cells 
oblong rather than oval. 

It is quite the usual thing for a spherical species of Chlamydomonas to have 
an oval or oblong variation. Thus of Chi. glolmlosa Perty as commonly found 
here (diam. 14-25/x) Wille, op. cit. says "Die Zellwand der Zoospore ist kugelrund 

oder g^chwach ellipsoidisch" I have myself figured that of Chi. monadina 

(Austr. Frw. Phytopl. These Proceedings, 1917, PL Ivi., f. 16). Compare also 



208 NOTES ON FRESHWATER ALGAE^ 

CM. maculata Playf. (These Proceedings, 1918, PL Iv., f. 15, f. 19 and f. 18) ; 
the last being obviously intermediate between the other two. 

An ovate form of this type is described below. 

Chlamydomonas taurangensis, n.sp. (vide infra) was present also. 

ii. — Examination of certain green fluffy growths. 

On February 13th, 1920 (summer) after heavy rains, my attention was 
drawn to certain green fluffy tufts adhering to grass blades in rainwater pools 
on a piece of meadow land previously quite dry. Minute observation of such 
growths often brings to light new, rare or interesting forms of life, besides 
affording much useful information. 

a. — In this case the green tufts proved to have as a foundation a mass of 
infantile filaments of Oedogonium sp. so unformed as to be hardly recognizable, 
except by their characteristic flecked chloroplasts and a slight enlargement of one 
end of certain cells. 

The Oedogonium cells began to form zoospores. One or two cells near the 
end of a filament twisted at right angles, the filament broke across and the eon- 
tents of the cell gTadually drew out. Occasionally the inner portion of the 
cytoplasm lagged behind but was later absorbed. There was no fusion of the 
cell-contents, they remained disposed exactly as in the cell wdth a small central 
nucleus plainly visible, and a small pyrenoid at each end. In general, the shape 
of the zoospore is more or less oval or long oA'^ate, but a cylindrical form was 
also noted. In front it is produced into a large ciliated papilla of clear homo- 
geneous protoplasm. jSTo stigina, no pulsating vesicle nor chlamydomonadine 
structure observed. Size of zoospores : — long. c. 30, lat. 15/a. They soon settled 
down to form new filaments. 

In a gathering of Cladophora sp. made at Tauranga, N.Z., on January 3rd, 
1922, the subsequent development of these zoospores was noted. They form 
Cliaracio psisAJike resting cells, from the apex of which young Oedogonium fila- 
ments push out. Dimensions of resting cells: — long. 20-40, lat. 8-18/x.; they are 
more or less oval, adhesive disc distinct, chloroplasts finely granular, sometimes 
obscurely divided into four parts, no pyrenoids visible. When devoid of con- 
tents they cannot be distingniished from CJiaraciopsis. 

h. — Entwined among the Oedogonium filaments, Leptothrix threads in quan- 
tities were busily breaking up into Spirillum volutans. 

c. — Chlamydomonas glohosa, vegetative cells in quantity attached to an al- 
most invisible hyaline membrane; cells rapidly becoming motile; diam. 3-10/x, 
globular. Var. maculata mihi also present, as well as Chi. Snowii Printz {see 
above); dimensions: — long. 10, lat. 6/^, with huge pyrenoid (diam. 3^), half the 
breadth of the cell. 

d. — Chlamydomoxas mucicola Schmidle. 

Algenflora d. Scliwarzwaldes u. d. Oberrheins, vi., 1897, p. 17-19. T. ii., 
fig. 4-8: rf. AVille. Gatt. Chlamyd. (Alg. Notizen, xi., p. 136, T. iv., fig. 6). 

Yar. rotundata, n.var. 

Fonna cellulis sursum rotundatis, papilla nulla; stiiiinate distineto, paullo 
supra raediam; pyrenoidi maxima ad latus disposita. Long. 9, lat. 4^/a. 

Differs from the type in being rounded above, not acuminate, and without 
tlie apical papilla. A stigma too is distinctly visible a little above the middle 
and the pyrenoid very large, f|uite half the width of the cell. It does not lie in 



BY THE LATE G. I. PLAYFAIR. 209 

the lumen of the cell (though it may appear to do so when the latter is in a 
certain position) but is placed laterally. This Chlamydomonas, which I now 
observed for the first time, was present in quantity, cells both motile and non- 
motile. The body is rather irregular in outline, varying from elliptic to long- 
ovate, often with uneciually arched sides, probably the result of longitudinal 
self -division. Chloroplast bag-shaped, with pyrenoid median and lateral; and the 
nucleus at the hinder end of the cell. Schmidle gives as dimensions : — long. 6-8, 
lat. 3-4^. 

A very similar species is CM. Kuteinihowi Goroshankin (Morph. u. System. 
d. Chlamyd., 1891, p. 117, T. ii., fig. 9-13) which has a thin laminar chloroplast 
and distinct median stigma, but is double the size. I hesitated for long whether 
or not to place our form under Goroshankin's type as the latter has priority. I 
was only decided by the dimensions. 

e. — Spondylomorum quaternarium, var. rostrattim Playf. (These Proceedings, 
1918, p. 526, PI. Ivi., fig. 16, 17). Minute 8-celled coenobia noted, only 16/x in 
diameter. 

/. — Euglena pisciformis var. granulosa, n.var. — {See Note xix.. p. 223.) 

iii. — Ulothrix qiiaternaria, n.sp. (Text-figs. 1-5.) 

Ulothrix filamentis longam massam cericeam saturate viridem formantibus. 
Fila latitudine 7-9 (rarius 10-11) ^ ad septa baud constricta; apicibus acute 
eonicis; cellulis plus minus quadratis (7-12^ alt.) diametro 1-1-| plo longioribus; 
vel cylindraeeis (17-28/* alt.) diametro 2-4 plo longioribus. Chloroplastides parie- 
tales, pyrenoidibus 2-4 (rarius 6) transverse geminatis dispositis (plerumque 4 
per quadrum ordinatis). Zoosporae globosae, ovatae, vel fusiformes, long. 8-21, 
lat. 7-10;*; flagellis 4. 

Forma angustissima ; filis latitudine IjH, cellulis cylindraeeis (10-30/* alt.) 
diametro 21-6 plo longioribus, polos adversus pyrenoidibus singulis. 

This plant I noted first in January, 1918 (summer) forming deep green 
silky skeins in a cement water table near the Post Office, Lismore. There were 
yards of it, and it could be taken up by the handful; the water was quite warm 
to the touch. The filaments were Sjx in diameter (rarely 10/*) not constricted 
at the septa; cells varying from cj[uadrate (12-15/* alt.) to cylindrical (generally 
21-25/t alt.) 2^-3 times longer than the diameter. In general outline it very 
much resembled Microspora stagnorum (Kiitz.) Lagerh. in Hazen, Vlothric. and 
Chaetoph. of the U.S., PI. 24, f. 13. Besides it answered very well to the descrip- 
tion of Ulothrix subtilis Kiitz. (in Rabhenhorst, Fl. Eur. Alg., iii., p. 365) now 
considered a species of Stichococcus. However, on mashing up some filaments 
on a glass slip, no H-pieces showed in the broken cells which also refused to 
dissociate at the septa as in Sticlwcoccus. The plant therefore seemed to be a 
species of Ulothrix. Apart from the size and shape of the cells its chief charac- 
teristic is the formation of a large number of pyrenoids, generally 4, arranged 
in a more or less regular square near the centre of the cell. 

Soon afterwards I found it again in company with a lemon-yellow slime 
fungus scraped off the wooden supports of a tank and almost aerial in its 
surroundings, being merely wetted now and then with rain water. The filaments 
were again 8/t broad, cells alt. 14-16/*, 2 pyrenoids in each cell placed transversely 
and becoming 4 by division. This formation of pyrenoids would therefore ap- 
pear to be a specific character. 



210 XOTES ox FRESHWATER ALGAE, 

A little later the plant came to hand in quantity from a cement water-table 
in Carrington Street in the waste water from the Electric Power House. Fila- 
ments then 7-9yu, broad with acutely-pointed conical tips were noted. Cells 
either quadrate (alt. 7-12^) or cylindrical (alt. 17-28^) ; pyrenoids as before. 

This sample being brought in late in the evening, was left uncovered on the 
lid of a tin all night. Next morning by 9 a.m. the exposed portion was actively 
engaged in forming zoospores. A thin-walled globose cyst developed in the cell 
and burst through the cell-wall on one side. Within this the zoospore formed 
(Text-fig. 1) and later broke forth, but often the cell-contents simply forced a 
way at once through the cell-wall and issued as a zoospore. The process took 2-3 
minutes. I endeavoured to observe the origin in the cell of the pulsating vesicles 
but without success. 

From each cell, whether long or short, one zoospore resulted. They are 
globose (diam. 8-10/a), ovate (15-19^ long by 7-10^ broad), or shortly fusiform 
(long 16-21^, lat. T-Sfi) gradually becoming more or less linear (Text-figs. 2-5). 
The contents are arranged exactly as in the cell but the pyrenoids scattered here 
and there, no Chlamydomonadine structure, flagella 4, pulsating vesicles 2 alter- 
nating, no stigma. The shape is not permanent; even while motile the zoospores 
begin to gTow, at once passing rapidly from one shape to another and becoming 
longer, narrower, and more attenuate behind. 

Involved among the UT-othrix filaments were quantities of the following 
species. 

Chlamydomonas Snowii, var. ovata, n.var. (Text-fig. 6.) 

Forma ovata, sursum plus angustata, f route modice acuminata. Long. 8-10, 
lat. 5-6/x. 

One of the congeners of CM. (jlohosa Snow, some of which are generally to 
be met with in gatherings svieh as this {see Note i.). There the pyrenoid was re- 
markably large for the size of the cell, probably preparatorA? to longitudinal self- 
division of the cell, and in those under review here, many of the pyrenoids had 
actually divided. Cells somewliat obtuse in front are included in the type; these 
are more or less pointed. Chloroplasts bag-shaped or campanulate; stigma 
median. 

There is no essential difference between thin laminar chloroplasts, those which 
are bag-shaped, and those which are campanulate; they are merely three degrees 
of development. 

iv. — Iiivestif/ation of some green ivater from a duck-pond. 

On October 10th. 1921 (spring), being then at Tauranga, N.Z., a small 
quantity of green water was brought in to me from a neighbouring duck-pond. 
Tlie colour turned out to be due to a small Chlamydomonas in the PaimeZk-state 
from the fine structureless mu<'us of which the cells were just breaking forth. As 
I know of no previously described form witli whicli it can be identified T have 
named it after tlio township. 

CHTiAMYDOMOXAS 'I'.U'RAXGENSIS, U.Sp. 

Cellulae parvae. oblongae, latitudine paene exacto duplo longiores, sursum 
et infeme rotundatae (intordum longeovatae vel inaequales, pone subtruneatae) ; 
latei'iVms arcuatis. saepe iiiaequalibus. Cliloroplastis pavietalis bui'sao-formis; 
stigmafe hemi-sphaerico, distincto, mediano. Pyrenoidis magna, media cellula ad 
latus disposita. 



BY THE LATE G. I. PLAYFAIR. 211 

Long. 13-20, lat. 7-10/a. Tauranga, N.Z. 

The cells were in immense numbers, both motile and non-motile. Small, and 
in general very irregulai-, they gave evidence in my opinion of having been formed 
by longitudinal self-division, one side being more arched as a rule than the other. 
Two of the largest (19 x 10/a) that I noted lying side by side in the mucus, were 
however perfectly formed, in outline elliptie-oblong with evenly arched sides and 
broadly rounded ends. ChloroiDlast parietal, bag-shaped. The chief characteris- 
tic of the type is the arrangement of the stigma, nucleus and pyrenoid trans- 
versely at the centre of the cell. Pyrenoid very large, quite half the breadth of 
the cell, and though sometimes appearing central, is really lateral. Irregiilar 
cells have a tendencj^ to ])e subtruncate below. 

I met with this form once before, in company witli Chi. glohosa, at Lismore, 
N.S.W. {see Note i.). 

Examined again on October 18th, the phial having been kept for eight days 
exposed to a strong diffused dajdight, the mucous stratum still showed quantities 
of the same type, mostly non-motile cells without stigma. Clear evidence was pre- 
sent of their origin as autospores from a minute spherical cell of lO-lS^a diameter 
which is a large size of Chi. glohosa. These with 4 autospores side by side were 
to be noted everywhere. Well-developed specimens of Chi. glohosa (diam. 8-9 fi) 
as 4 autospores in a mother-cell were present also. 

But the most interesting item of the contents was a bizarre form of Chlaniy- 
domonas to which I have given the name of 

Chlamydomonas novaezealaxdiae, n.sp. 

Cellulae modiee ovatae vel oblongae (interdum etiam subquadrangulares), in 
ambitu valde inaequales, permagnorum gTanulorum amylaeeorum per membranam 
delieatissimam projicientum seriebus ternis formatae; granulis plerumque rotun- 
datis, anterioribus medianisque maximis. Chloroplastis vix visibilis; stigiuate 
nuUo nee pyrenoidi; flagellis tenuibus, brevibus, binis. 

Long. 5-8, lat. 3-6/x. Tauranga, N.Z. 

The cells were, in general outline, ovate or oblong, sometimes verging on sub- 
quadrangular. They are chiefly characterised by 3 series of very large rounded 
paramylon granules projecting through the very delicate membrane which can 
only be detected as a faint tag at the apex of each granule. End view shows 
that these are all round the cell. The two u]3per series are the most distinct, but 
the shape is very variable. A faint colouration indicates the presence of a 
ehloroplast, but stigma and pyrenoid were alike absent. I thought I detected 
contractile vacuoles and a minute central nucleus; two short delicate flagella quite 
certain. 

Quantities were present, both motile and non-motile, the latter often arranged 
in sets of four. But the point most worthy of remark is that in the original 
gathering not a cell of this type ivas present, so that we have here the production 
of a new species from a perfectly different form by repeated self-division and 
formation of autospores under pathological conditions. 

V. — On certain forms of Chlamydomonas. 
Chlamydomonas stellata Dill. 
Die Gatt. Chlamyd., 1895, p. 17, T. v., f. 31-36; of., Wille, 1903. p. 134. T. 
iv., f. 4. 

Form without anterior papilla, ehloroplasts more attenuate inwardh^, stigma 



212 



NOTES OX FRESH WATLOn ALGAE, 




Text-fibres 1-15. 



1-5, Ulothrix quaternaria, n.sp. (1, x 800; 2-5, x 1100). 6. Chlamydomonas 
Snoioii, var. ovata, n.var. (x 2400). 7. Chlorogunium elegans, n.sp. (x 1100). 
&-12. Gomum pectorals. Development of the cells (Palmella-state) to form a 
16-C'oenobia family. 13a.-h. Ophioeytium capdtatum var. longisjnnum. 14, 14a. 
Euglena australiea, n.sp. (x 75(5). 15. Euglena australica, var. gibberosa, n.var. 



BY THE LATE G. I. PLAYFAIR. 213 

not observed; syn. CM. rotula Playf. These Proceedings, 1918, p. 517, PL Iv., 
f. 12. 

Wille gives: — Long. 18-20, lat. 10-13|U,; ours are 19-21 x 15^, pyr. diam. 6/x- 
Gloeocystis-states are Asterococcus superbiis (Cienk.) Scherffel, 1908, Ber. d. 
Bot. Ges., Bd. xxvia (c/. Pascher, op. cit., H. 5, p. 50, f. 302). Oocystis rotula 
Playf. {antea, 1916, p. 130, PI. vii., f. 31) ; Asterococcus limneticus G. M. Smith, 
1918, Trans. Wis. Acad., p. 627, PI. 10, f. 3-6; Wisconsin Phytoplankton, 1920, 
p. 104, PL 20, f. 7-10. 

Chlamydomonas Steinii Goroshankin. 

Morph. u. System, d. Chlamyd., ii., 1891, pp. 112, 113, T. iii., f. 1-8, 29, 30; 
Wille, 1903, p. 134, T. iv., f. 5. 

Form sometimes with a minute but distinct (I/a alt.) anterior papilla, and a 
pyrenoid at each end. The apical one seems to be rather an agglomeration of the 
anterior ends of the chloroplasts. Plenty noted in a floating Euglena scum. 
Goroshankin's dimensions are: — Cell. long. 18-30 (generally 24yu,). 

Cell. long. 21-23, lat. 12-13^. Lismore. 

Chlamydomonas obsoura^ n.sp. 

Cellulae oblongae, ubique rotundatae; lateribus levissime arcuatis; fronte 
papilla hemisphaeriea distinct a; chloroplaste crassa bursae formi, cellulam paene 
complente; nucleo mediano; pyrenoidi nulla (in speciminibus notatis) ; stigmate 
hemisphaerico mediano distinctissimo. 

Long. 12-15, lat. 7-8^^. Lismore (N.S.W.) ; Tauranga (N.Z.). 

A small form found in a wayside waterhole draining a swamp. Its general 
outline is oblong with broadly rounded ends, but sometimes verging on ovate from 
being narrowed anteriorly. A very distinct hemispherical papilla in front, at 
least in the motile specimens. Cytoplasm minutely granular, dense; ehloroplast 
massive, almost filling the cell, leaving merely a narrow and sometimes irregular 
protoplasmic core. Nucleus central; no pyrenoid noted; stigma hemispherical, 
median, very distinct. 

Gloeocystis-state observed: — subglobose, 16-eelled, diam. 100/^. Cells without 
apical papilla, pyrenoid or stigma, but with the same dense finely granular cyto- 
plasm and massive ehloroplast. 

Var. ovata^ n.var. 

Cellulae ovatae, pone latissime rotundatae, fronte acutae; lateribvas valde 
arcuatis. Cetera ut in forma typica. 

Cell. long. 12-15. Tauranga (N.Z.). 

Found with the type, both motile. This is a very decidedly ovate form, 
acute in front, broadly rounded behind, sides well arched. Breadth doubtful. 

vi. — Development of Chlamydomonas from resting cells. 

On July 24th, 1915, after several days' heavy rain my attention was directed 
to a small green flake about the size of a halfpenny floating on the surface of a 
muddy roadside pool. It proved to be composed of infantile non-motile forms 
of some species of Chlamydomonas. There were all shapes and sizes, lying in 
an irregular plexus, held together on their sides by a thin membranous pellicle. 
Three sorts might be noted, viz: — 

a. — A comparatively few fairly well-developed cells, oval or ovate, long. 13- 
20, lat. 8-14^, showing the usual bell-shaped ehloroplast, very pale green and 



214 NOTES ON FRESHWATER ALGAE^ 

homogeneous, with sometimes a pyrenoid or pyrenoidal loeellus. The flagella 
might often be observed feebly fluttering. 

6.-^ A much larger section consisting of smaller long-elliptic cells about 10 x 
4/*. Chloroplast laminar, very slightly coloured indeed. 

c. — A crowd of smaller cells very irregular in shape, up to 6 x 3^ in 
dimensions. Very often just a ti'ace of chloroplast with barely any colour. 

In no cell was any stigma noticeable, but even in the small ones contraetiLe 
vacuoles were seen working. Even as I watched the cells under the microscope 
one or other would become active — a few slight movements — some more decided 
wriggles — and it broke free. It was curious that it was always the minute and 
often the more irregular cells which became active, very rarely, if ever, the well- 
developed ones. This is quite the rule auioug flagellates. 

The stratum examined was not a Pa/weZkt-state where the cells are immersed 
in mucus, but a tough, though delicate, papyraceous membrane with the cells 
affixed by their sides. This membrane I believe to be formed on the drying up 
of the pool by very fine flocculent matter floating suspended in the water, while 
the Chlamydomonadine cdls present evolve swarm spores which survive the 
drought as minute resting cells adherent to it. 

Consulting, as I do, from time to time, Wille's most useful monograph 
"Uber die Gattang Chlamydomonas" I have been struick by the regularity of the 
reference to the zygote in every species. Nevertheless, though I have done a good 
deal of work searching just the places in Avhich zygotes might naturally occur, I 
have never met with a single specimen, and, from the material I have found, 
have been forced to the conclusion that some alternative method of development 
such as I have suggested, must exist. 

vii. — The Genus Chlorogonium. 

CHLOROGO>aUM EUCHLORUit Ehr. 

The form that seems to have been generally accepted in Europe as the type 
is Dajigeard's figure reproduced in Chodat, Algues vertes de la Suisse, Bd., i., H. 
3, p. 40. It is very broadly fusifoim, merely acuminate behind, acutely rounded 
and only slightly rostrate in front. Tlie dimensions. I.e., p. 139, are given as: — 
long. 30-50/a; laf. 8-12/x. 

The nearest approach to such a form that I have observed was a single 
specimen from the Botany Water Reserve, acuminate at each end, without rostrum 
or tail. It was alive and active in spite of the fact that the cytoplasm was 
divided into 8 oval autospores. Long. cell. 3T), lat. VZ^i. 

Chlorogoxium elegans, n.sp. (Text-flg. 7.) 

Cellulae graciliores, fronte acute-rostratae, pone aeutissimae, sensim sensim- 
que attenuatae. 

Long. 36-55, lat. (i-lO^u,. Botany. Lismore. 

More slenderly fusiform than Chi. euchlorum and more giadually attenuate 
to each end. In front more acute and rostrate, behind very acutely drawn out 
almost into a spine. Chloroplast a thin parietal lamina without pyrenoids. 
Stigma orange-brown, wick-shaped when sliowing at any i-ate (probably, hoAv- 
ever, a delicate flat circular disc in face view) a little in front of centre. A pair 
of conti'actile vesicles about the anterior fourth. Nucleus a little behind the 



BY THE LATE G. 1. PLAYPAIR. 215 

centre, elosc4y surrounded ])y a locellus. Cytoplasm occasionally retracted from 
the tail. 

* Chlorogonium Stbinii^ nom. uov. 

As in Stein, Naturg. d. Flag., H. i., T. xviii., f. 6. The figure Avorks out 
at:— Long. 123, lat. 12^.^ 

I came across a pure culture of this large species in an iron saucepan in a 
fowl-yard. Both active and non-motile cells were present in abundance, the latter 
(as is usual in both Volvocine and Euglenoid flagellates) fixed in clumps by the 
anterior end of the cell to pieces of flocculent debris where they had evidently 
developed from - resting cells. Under the microscope the vegetative cells (Con- 
tinued to become motile. 

Compared with the breadth they were very long; in front produced into a 
long rostrum, posteriorly drawn out into a very acute tail devoid of cytoplasm. 
Two flagella proceed at right angles to the rostrum just below the extreme tip 
and a pair of small pulsating vacuoles indicate what may be considered the base 
of the rostrum. About the centre of the cell a small nucleus surrounded by a 
locellus almost as wide as the cell itself. I did not note any stigma. Chloro- 
plast a delicate parietal lamina with 8 small pyrenoids arranged in a longitudinal 
spiral above, and 8 below the nucleus. 

Cell. long. 62-93, lat. 7-10; lat. rostr. 1^/x. Lismore. 

Chlorogonium minimum Playf. 

New and rare frw. Algae. These Proceedings, 1918, p. 521, PL Iv., f. 26. 
Cell. long. 30, lat. 2-3;a. 

Var. gurtujM, n.var. 

Cellulae breviores, et fronte atque pone minus protractae. 

Cell. long. 24, lat. 4ju,. Lismore. 

A shorter more immature form, less produced both in front and behind. 
Only a very delicate homogeneous parietal chloroplast noticeable, and a very 
distinct orange-brown stigma somewhat forward of the centre. Almost certainly 
a developing stage of the type; plenty were present in a Euglenoid scum from a 
roadside pool, March 12th, 1920, at Lismore. 

Var. OBESUM, n.var. 

Cellulae Ijrevissimae, obesae, inaequaliter longe-ellipticae, et fronte atque 
pone obtusae; uno latere areuato, altero deplauato. 

Cell. long. 12-14, lat. 3(x. Tauranga, N.Z. 

From a duck-pond (Oct. 18th, 1921). An extremely lively plump little growth 
form, distinctly green. Chloroplast very delicate, parietal; two flagella; stigina 
in the anterior portion easily noticeable; no other details. 

With this form was noted what looked like the Palmella-state : — narrowly 
cylindrical cells in pairs, pale green homogeneous contents; cell. long. 8, lat, 2^. 

What were pi'obably sessile vegetative specimens of the zooid, affixed to 
some animalcule 1 obtained from a pond in the Botanic Gardens, Sydney (June 
30th, 1910). Cells linear-elliptic, long. 12, lat. 3fi; ehloroplasts bright green; 
they became motile on being broken loose from the host. 



216 XOTES ox FRESHWATER ALGAE^ 

viii- Development of Pandorina and Goiiium. 
Pandorina. 

In September, 1909, I obtained quantities of small freshwater Algae by 
shaking out heads of Elodea and Myriophyllum growing in a dense mass close 
to the surface of the water in a tank at the Botanic Gardens, Sydney. Among 
them were numbers of Pandorina morum, exhibiting the various stages of develop- 
ment by subdivision of a single resting cell. All the forms were noted in a 
single drop beneath the microscope at one time and might therefore easily be 
compared. 

a. — Undivided resting cell. The smallest seen were diam. 10-14yu, surrounded 
hy hyaline mucus 2^ wide. Cytoplasm very pale gi'een, finely granular. 

h. — Undivided resting cell. Larger size, diam. 16-18ju,, ring of mucus 2/x, 
wide. Cytoplasm a shade darker green, finely granular. 

c. — Cell subdivided into a globular coenobium, non-motile; diam. about 24- 
30^. No flagella or pores visible. Cytoplasm still very pale green, finely 
granular with darker grains here and there. 

d. — Oval coenobium, non-motile, long. 31-38, lat. 25-34^. No flagella. but 
pores visible. Cytoplasm darker green. 

e. — Larger oval coenobium, non-motile, long. 48-52. lat. 44^; no flagella, but 
pores visible. Cytoplasm darker green. 

Compare P. morum var. tropicum Playf. (These Proceedings, 1915, xl.. 
Part 2, p. 336, PI. xliv., f. 18). It seems to me quite unreasonable to suppose that 
these irregular polygonal cells remain so and I feel more inclined than ever to 
believe that they become regularly spherical and form Eudorina elegans. 

Pandorina morum var. coelastroidea_, n.var. 

Coenobium minutum globosum, Goelastro microporo consimile; e eellulis 
minimis paene ajppressis exstruetum; locello centrali sphaerico (ut videtur) in- 
structum. Cellulae in quincuneem (6-|-l) dispasitae, e vertice visae hexagonae, 
e latere euneatae fronte rotundatae. 

Coenob. diam. 23;^; cell. diam. 4; loeell. centr. diam. 8^, Lismore. 

An unusual form which, owing to the extreme tenuity of the investing mucus, 
looks Like a flagellate specimen of Coelastrum microporum. The edge is crenate 
as the outer margin of the cells is rounded. The cells are very small (diam. 4^) 
almost regularly hexagonal from above and neaily appressed, arranged quincun- 
cially 6 round 1. There appears to be a central globular loeellus Sjx in diameter. 
Flagella distinct and the pores indicate the presence of a mucous investment. 

Development of Goniuni. 

In the gathering mentioned above, I was able to trace all the vegetative 
stages of Goriium soeiale from a single resting cell. But how did tlie resting cell 
originate? Unfortunately, I did not obtain their dimensions ])ut the smallest 
may be taken as diam. 5^. 

a. — Undivided pale green cell with very narrow mucous investment. 

&.— Cell-division in one direction; cytoplasm pale green. 

c. — Cell-division in a second direction, thus forming 4 angular cells. 

d. — Cells become globular (see my remarks under Pandorina). 

e. — Cells a diameter apart. Later they develop flagella and tlie coenobium 
becomes motile. Sometimes the cells take on an oval or ovate form. Cf. Chodat, 
Algues vertes, p. 148, f. 73. 



BY THE LATE G. I. PLAYFAIR. 217 

ix. — Development of Selenasti'um and Coelastrum. 

In the same gathering' (September, 1909, supra) was a series of cells with- 
out mucous investment showing the development of Selenastrum Bibraianum from 
a single minute globose cell. 

a. — Minute globose cell, diam. 'q^. 

&.— Larger globose cell, diam. 12yu,; chlorophyll regularly diffused. 

c— Globose cell with chlorophyll Innately disposed, diam. 12^. 

d. — Cell growing out laterally in a lunate manner, long. 20, lat. 10^. 

e. — Still more lunate, apices shaiply pointed, diam. from point to point 24ja, 
lat. max. 10/x. 

/. — A variant, long. 32, lat. max. 14;^. 

g. — Fully developed cell, long. 24yu., lat. max. 10/^. 

In nature the mature cell divides along the plane of the apices and body. 
The new cell twists at right-angles to the first and remains attached by the centre 
of the body at the back. Two cells are often to be seen like this. 

Coelastrum. 

A gathering at Lismore (December 13th, 1912), among other things, supplied 
me with material showing how small coenobia of Coelastrum are often built up 
from a single cell hj a process of self-division in two directions. 

a. — Single cell, diam. 5/a. 

b. — Self-division in one direction, long. 10, lat. bfx- 

c. — Self -division in a second direction; diam. coenob. IG, cell. 8/a. 

d. — An arched plate of six cells (5-]-l) ; diam. coenob. 22, cell. 8^. 

Cells beautifully deep transparent green; a large pyrenoid in each. 

Small complete coenobia in same sample. 

X. — Rejuvenescence in Gonium and Botryoeoceus. (Text-figs. 8-12.) 
On two separate occasions when making observations with the microscope I 
was so fortunate as actually to Avitness this process. In the first instance it was 
in Gonium pectorale. 

The matrix of a eoenobium having become old and stiff, it had evolved a new 
and larger mucous plate attached to the old eoenobium. The cells had lost the 
flagella and become vegetative. One by one with an explosive action they left 
fheir old settings and look up positions at even distances apart in the fresh 
mucous cvishion. The resultant condition was Palmelloid. 

Each cell undergoes self-division, first in one direction then in a second 
transverseh', until finally a family of 16 perfect coenobia is produced (Text- 
figs. 8-12). Such I have noted on more than one occasion and once was so for- 
tunate as to obtain such a family showing all the stages of development. But 
the cell-division as observed in nature does not altogether follow the lines laid 
down by Chodat in Algues vertes, p. 42. 

BOTRYOCOCCUS. 

The ease of Botryoeoceus Braunii was similar. The cells were oval and 
seemed to be longitudinally divided, each being surrounded by a layer of stiff 
mucus. Attached was a cushion of newly formed clear hyaline mucus into which, 
as I watched, the cells bursting from their investment passed with a sudden and 
explosive jerk, becoming young and globular and settling down at about equal 
distances apart. The newly formed condition was Palmelloid and the old mem- 
brane fell away. Each cell should now divide to form a new eoenobium. But 



218 ■ KOTES ON FRESHWATER ALGAE_, 

of what ehax'aeter? For it is evident that Botryococcus is a pathological condition 
of some othei" alga caused by the drying up of its habitat. 

xi. — ARTiiiU)OcysTis ellipsoidea W. & G-. S. West. 

Welwitsch's African Algae, Journ. Bot., 1897, p. 238, T. 370, figs. 1, 2. See 
Wille, Conj. u. Chlor., in Engler u. Prantl., Die naturl. Pfianzenfamilien, 1909, 
p. 39, fig. 19r. 

This organism is probably the daugbter-coenobium of Volvox tertius Meyer. 
There is an oval form of that species, and in it the daughter coenobia are 
generally oval also, and whether enclosed or free, the cells of the latter are often 
represented by a fragmented cliloroplast without flagella. Compare my notes in 
Frw. Algae of the Lismore Distr. (These Proceedings, 1915, pp. 340, 341, PI. 
xliv., fig. 2, 3) under Volvox tertius var. tesselatus. These daughter-coenobia, 
with parietal laminar chloroplast fragmented in a tessellate manner, are common 
enough in tropical and subtropical districts. 

xii. — Note on the horns of Pediastrum. 

Dealing with the genus Pediastrtrm (in Chlorophyeeae, ii., Heft 5, of 
Pascher's Die Siisswfl. Deutsch. 1915) Brunnthaler remarks, p. 89, that, according 
to Lemmermann, these processes (the horns) are hollow and open (fig. 50) and 
allow a thin filament of plasma to stretch forth. Lemmermann's observation re- 
fers to P. simplex. Chodat {op. cit., 1902, p. 227, f. 152a) has remarked the 
same circumstance in P. duplex v. reticulatum. 

In 1909 1 noted it also in living specimens of P. duplex var. clathratum A.Br, 
obtained from the plankton of the lagoon in Parramatta Park. The filaments are 
extremely tenuous, sometimes as many as three from each horn, and occasionally 
vary their direction. 

A. Braun however (Alg. unicell., 1855, p. 67, footnote 1) traces this observa- 
tion back to qiaite early microscopists. He quotes Turpin (Mem. du Mus. d'hist. 
nat., x\d., 1827, p. 320) as saying: "Lorsqu'on observe eette production dans un 
lieu chaud, il n'est pas rare de voir plusieurs des cornes muqueuses lancei', de 
leui-s extremites une poussiere de globules," Turpin therefore must have con- 
vinced himself that the horns were hollow. 

Braun goes on to say {ibid., p. 67) that Ehrenberg, relying on Turpin's 
observation, holds the opinion (Infus., p. 155) that the horns are perforated hy 
small orifices. 

I watched long and carefully for Turpin's "poussiere de globules" but never 
saw anything pass from the horns. 

xiii.- — Ineffigiata neglecta W. & G. S. West. ^ 

This plant is generally found as small irregular masses of a brownish-green 
or yellow-gieen colour out of pond weeds. I obtained it also from the Sydney 
Water Supply in large conglomerations, the pieces being connected by what were 
evidently strings of dried mucus. The cells are said to be disposed side by .side 
as in Botryoeoccwi but within tlie investing membrane. 

I gathered a peculiar and unusual form of this plant from Canley Vale in 
some quantity. The membriine of each piece seemed to be covered with per- 
forations througli each of which i)rotruded very slightly a low rounded papilla 
of hyaline transparent mucus. From the apex of these proceeded a very long 
delicate pituitous thread in length equal to twice or thrice the diameter of the 



BY THE LATE G. I. PLAYFAIR. 219 

mass. Those threads Avere of (luite a different character from the coarse mem- 
branous strands often found joining these masses, and under a high magnifieation 
showed no signs of being composed of Bacterium cells. The interior of the mass 
seemed to be replete with a number of large hyaline refringent bodies; one was 
distinctly visible through a rent in the membrane and by putting the microscope 
a little out of focus the position of others was indicated (as is often the ease) 
by a refringent glow. Quantities of the usual minute pieces of the plant were 
observable in the same gathering: — green, yellow-green, orange and even brick- 
red in colour. One mass measured 50^ long and broad. 

xiv. — On certain species of Oocystis. 
OocYSTis BoEGEi Snow. 

Plankton of L. Erie, Bull. U.S. Fish. Comm., 1903, p. 379, PL 2, fig. vii. Syn. 
O. gigas var. Borgei Lemm. ; Oe Novae Semliae var. australica Playf. (These 
Proceedings, 1916, p. 123, PI. vii., figs. 13-17. I have lost my copy of Snow's 
memoir, but the identification is certain, as my figures agree perfectly with that 
given by A. Pascher in Slissw. flora Deutsehlands, H. 5, "Anhang," p. 234, f . 38. 
The dimensions for American specimens are : — Long. 9-17, lat. 9-13yu, ; European : 
Long. 13-17, lat. 9-13;^; Australian :— Long. 13-20, lat. 10-14;*. Ours are from 
the plankton of the lake in Parramatta Park and from the Richmond River at 
Lismore. The chloroplast is laminar, parietal, irregularly fragmented. 

Oocystis crass a Wittrock. 

Syn. 0. crassa var. Ostenfeldii Playf., forma, These Proceedings, 1917, p. 
839, PI. Iviii., f. 1; 0. crassa var. elongata Playf., pro parte, ibid., p. 839, PL 
Iviii., f. 2. 

The original figure of 0. crassa being so irregnilar in shape and of such doubt- 
ful characteristics, the form generally accepted as typical is broadly biconvex, 
not incrassate or tuberculate at the poles^ with 8 ehloroplasts each with a pyrenoid. 
Our specimens as figured agree with G. S. West's in Brit. Frw. Algae, p. 227. 
In regard to size W. and G. S. West give long. 18-24, lat. 13-15.5/x for British 
specimens; European are recorded at: — Long. 14-26, lat. 10-20^; ours are long. 
20-25, lat. 13-15/i. 

0. crassa var. Ostenfeldii Playf. has 4 rounded discoid ehloroplasts arranged 
generally longitudinally. My fig. 3 (PL Iviii., I.e.) cannot be included under 0. 
crassa, but had better be considered as 0. solitaria Wittr. though it does not 
possess the polar incrassation within the membrane, characteristic of that species. 

Oocystis lismorensis, n.sp. 

Cellulae latissime ovales, ad polos quam levissime aeuminatae non autem 
apiculatae vel incrassatae; chloroplastidibus disciformibus vel plusminus in- 
aequaliter polygoniis (c. 6-7 utrinque visibilibus) vel circulatis minoribus (diam. 

4-5;a). 

Dimensions 43 x 35, 38 x 31, 36 x 30, 31 x 25yii. Lismore. 

A form rather difficult to place otherwise than by making it a distinct type. 
It is far too small to arrange under Eremosphaera {cf. Erem. vdridis var. 
acuminata Playf., antea, 1916, PI. ix., f. 3). 0. gigas Archer is oblong, not oval, 
with uninterrupted outline, as also is /. minor W. West. The cells in this species 
are very broadly oval, the least bit acuminate at the poles, but neither incrassate 
there nor furnished with an apiculus or papilla. The ehloroplasts are some- 



220 Is'OTES ON FRESHWATER ALGAE^ 

times large, irregularly polygonal, 6-7 a side, or smaller and circular. The axial 
ratio of the cells measured averaged 1 to 1.22. 

XV. — On two species of Ophiocytium. 

Ophiocytium capitatum var. longispinum (Mob.) Lemm. (Text-tig. 13, a-h.) 

Hedwigia, Vol. 38, 1899, p. 32, PI. 4, figs. 21-25. 

Syn. Reinschiella longispina Mobius; Schroder ia helonopJwra Schmidle 
(Beitr. z. Kenntn. d. Plankt., 1900); Centritractus belonopliorus (Schm.) Lemm. 
(Beitr. z. Kenntn. d. Plankt., ix., 1900, p. 273). 

In plankton from the lake in Parramatta Park and from Enoggera (Bris- 
bane) I observed a complete series of forms showing the growth from the initial 
spherical cell to a well-developed cylindrical one with 10 distinct band-shaped 
parietal chloroplasts. 

a. — Cell diam. Jfi, setae long. 26/x,- 

b. — Cell long. 16, lat. 7; setae 26^. Chloroplast a single parietal lamina. 

c— Cell long. 18, lat. 8; setae 33;^. 

d. — Cell long. 25, lat. 9; setae 57 fi. 

e. — Cell long. 36, lat. 8; setae 32^. 

/'. — Cell long. 99, lat. 8; setae 33^. 

g. — Cell long. 135, lat. 8; setae 35/^. 

Ji. — Very small specimen, pond plankton, from Lismore. Cell long. 11, lat. 
4, setae 23yu,, irregular in outline but showing 2 chloroplasts distinctl.y. 

Ophiocytium elongatum W. and G. S. West. 

Freshwater Algae of Burma, PI. xi., f. 11-13. 

The Burmese type is a remarkably long irregular serpentine filament. One 
-end is already beginning to coil, however, and f. 13 shows a specimen closely 
spiral. No length is given, but the thickness of the cell is stated to be 5-5^fx.. 
This Australian specimen if uncoiled would certainly be as long as the type, if 
not longer. 

Crass, cell 6^. Spira 90 x 30^^. Guildford. 

xvi. — Some rare forms of Scenedesmus. 
SCENEDESMUS Bi.jUGUs (Turpin) Lagerh. 

Nuova Notarisia, 1893, p. 158. 

Syn. Achnanthes hijuga Turpin; Scenedesmus bijugatus Kiitz. 

The form I figure here is the type as given by Kiitzing in Synopsis Diato- 
maeearum for his Sc. bijugatus and which he makes equivalent to Sc. bijugus 
Turpin. The cells, as he figures them, are oblong-cylindrical with rounded ends 
^nd sides flat or nearly so; 2, 3 or 4 parallel in a single series. I have only 
noted it once. It is a form evidently very rare, as another shape broadly oval 
(see Brunnthaler in Paseher, op. cit., H. 5, p. 164, f. 233) has obtained general 
recognition. See also G. M. Smith, Wisconsin Phytoplankton, PI. 37, f. 18-20. 

Var. DUPLEX, n.var. 

Cells divided transversely (not obliquely) exactly across the middle. Semi- 
t-ells slightly conical outwardly and one above the other. Each vertical pair of 
sernicelLs 16 x 10^.. From Fairfield, with tlie type. 

Var. IRREGULARIS, n.var. 

(.'elliilae interiores 2 transverse ordinatae. 



BY THE LATE G. I. PLAYFAIR. 221 

Coenob. (4-eell) 14 x 12, 18 x 10; cell. long. 6-9, iat. 4-5^. Botany Water 
Reserve. Fairfield. 

The two inner cells are arranged more or less at right angles to the other 
two. 

Var. ARCUATUS (Lemm.) W. & G. S. West. 

Syn. Sc. arcuaPus Lemm. in Ploner Forsch., vii., 1899, p. 17, T. i., f. 2-4. 
Se. bijugatus f. areuata (Lemm.) W. & Gr. S. West in Plankt. of some Irish 
Lakes, 1906, p. 106, PL s., f. 12-14. 

Coenob. (8-cell.) 12-20 long., 12-18 alt.; cell. long. 6-12, Iat. 2-6^,. Botany 
Water Reserve. Fairfield. 

Our specimens are small ; Lemmermann gives : — cell. long. 13-18, Iat. 7-9-^^. 
The slightly arched form of the coenobium is too insignificant a detail to found 
a species on; it is very often absent, the coenobium being flat. 

Var. DisciFORMis Chodat. 

AlgTies vertes de la Suisse, 1902, p. 213. "Cellules polyedriques par com- 
pression" — Chodat, I.e. 

Coenob. (8-cell.) 18-29^ long., 18-25 alt.; cell. long. 9-15, Iat. 4-7/^. Botany 
Water Reserve. Fairfield. Lismore. 

ScENEDESMUS Ralfsii, nom. nov. 

Cellulae ovatae alternantes, longo spatio inter se intermisso, in seriebus binis 
ordinatae. 

CeU. long. 16, Iat. 8/x. Collector. 

Syn. Sc. ohtusm Ralfs {non Meyen), Brit. Desm., T. 31, f. 16. Ralfs' 
figures do not answer to any of Meyen's types which, also, never having been 
properly described, cannot be accepted. The coenobium consists of 2 alternate 
series of ovate cells rounded at one end and acuminate at the other and separated 
bj'' a considerable interval. 

SCENEDESMUS ANTENNATUS Breb. 

In Ralfs' British Desmids, 1845, p. 222, T. 35, f. 27. 

Cell. long. 24, Iat. 5^. Botany V\^ater Reserve. 

Compare J. Brunnthaler Protococcales {op. cit., H. 5, pp. 163, 164, f. 211) 
who gives the size of the cells as: — long. 12-13, Iat. 2.5-4/x. See also Sc. Ber- 
nardii G. M. Smith, Wisconsin Phytoplankton, 1920, PL 38, f. 5-9, dimensions 
8-17 by 3-6^. 

SCENEDESMUS RENIFORMIS, n.Sp. 

Cellulae reniformes, inter se distantes, in seriebus binis ordinatae. 

Coenob. (8-cell) 23 x 15; cell. long. 7i, Iat. 4^. Lismore (327). 

A family of eight coenobia observed. A very rare type; the cells are per- 
fectly reniform, arranged in pairs one above another and facing opposite ways, 
each pair separated by an interval of about half a diameter. 

SCENBDESMUS CURVATUS Bohlin. 

Die Algen d. erst. Regnell Exp. L, Protoeoccaceen, 1897, p. 23, T. L, f. 41- 
44, 52. 

Verj^ rare everywhere apparently; noted here only from Botany Water Re- 
serve (Nos. 50, 95) ; plentiful however in the latter gathering. 

Coenob. (4-cell) long. 30, alt. 40; cell. long. 20-22, Iat. 7-9^. 

Coenob. (8-eell) long. 34-62, alt. 26-44; cell. long. 12-24, Iat. 4-10;^. 



222 NOTES ON FEESHWATER ALGAE^ 

SCENEDESMUS INCRASSATULUS Bohlin. 

Bohlin, Protoeoeeaeeen, op. cit., p. 24, T. i., f. 45-51. 

Cells shortly fusiform with a minute incrassation at each end. Very rare; 
seems to be known only from Brazil and Burma. Bohlin gives 17-24 by 5-8/a for 
the type. 

Cell. long. 12-20, lat. 5-6/^. Centennial Park, Sydney. Botany (Gardener's 
Road). 

xvii. — Notes on various species. 

UlOTHRIX AUSTRAIiICA, n.sp. 

Filaments at the base very narrow, rapidly increasing to 70yu, broad with 
cells transversely oblong, alt. 24-100;^. Upper portion of filaments 70-130/^ broad 
with palisade cells, alt. 18-24;t. Cells not constricted or inflated. Membrane Sfi 
thick a short distance above the base, soon attaining 10/a, at which figure it re- 
mains constant. Inner surface of the membrane marked with delicate irregular 
longitudinal and transverse lines (evidently ridges) which coalesce here and 
there. A single homogeneous parietal chloroplast filling each cell. No pyrenoids 
or amylaceous granules. 

Lismore (352). In company with the two following forms. 

Ulothrix zonata (Weber et Mohr) Kiitz. 

Fil. diam. 27-45, cell. alt. 11-38, membr. crass. 2-5 fi. 
Var. VALiDA (Nag.) Rabh. 

Fil. diam. 42-70, cell. alt. 11-32, membr. crass, 3-6/^. 

GOLENKINIA ACULEATA, n.Sp. 

Cellulae spherieae, aculeis paueis arcuatis brevibus munitae. 

Diam. corp. 20; acul. long. e. 7^. Sydney Water Supply (66). 

All my specimens of Golenkinia come from that source. This form is dis- 
tinguished by being armed with a few short, stout, arcuate prickles, about 16 
visible at a time. 

FLAGELLATAE. 

xviii. — Study of a red-brown surface film. 

The film was first noticed on March 12th, 1920 (Autumn), as a thin reddish- 
brown pellicle with dark purple-blaek ripple lines where the wind had piled it. 
Mottled green patches showed here and there. It turned out to be the matrix of 
a species of Euglena which I have not been able to identify with any published 
form. The film broke with a vitreous fracture and the locelli which had con- 
tained the vegetative cells were scattered irregularly throughout it. Circular in 
shape viewed from above (with delicate spiral lines showing), in side view they 
appeared elliptic so that the Euglena-cells were tabloid-shaped. I have named 
this species :- 

EUGLBNA AUSTRALICA, n.Sp. (Text-fig. 14.) 

Cellulae subcylindraeeae, lateribus plus minus parallelis, vel cylindrieo-fusi- 
formes; fronts attenuatae conicae, apicibus rotundatis; pone attenuatae conicae 
acuminatae, spina brevi aut nulla. Corpore metabolico. 

Long. c. 45-55, lat. 10-12/ji,. Lismore. 

When liberated from the locelli of the film, the Euglena-ceW is tabloid- 
shaped. The anterior and posterior portions grow out from the centre of the 
broad face on either side. At this stage the cell often becomes motile, stopping 



BY THE LATE G. I. PLAYFAIR. 223 

every now and again to knead itself into shape. I observed all these forms to 
pass from one to the other nnder the microscope. Each had a long flagellum 
and was very active. I distinguish this stage of development as: — 
Var. GiBBEROSA^ n.var. (Text-fig. 15.) 

Corpore transverse ovali; superne in projeetionem eonicam^ infeme in 
eaudam triangularem acutam producto; meitabolico. Cytojjlasma plerumque 
flavo-viridi, denso; chloroplastidibus min Litis digitis transverse dispositis; stig- 
mate subapieali. 

Long. c. 36, lat. e. SO^u,. With the type. 

It may be asked: why describe and name the stages of development in these 
metabolic forms? To prevent them from being erected into new species by other 
observers who come across them casually. For instance, I met with this bizarre 
form three years ago (April, 1917) in a gathering from another pond. It ex- 
hibited then the same characteristics: — tabloid-shaped body, conical anterior pro- 
minence and triangular tail-piece both hyaline. Cytoplasm dense, yellow-green; 
chloroplasms minute flecks disposed transversely; stigma in the anterior pro- 
minence distinct; nucleus not visible. Body metabolic. I had myself a mind to 
describe it as a new species. Other observers found species on the other 
transition forms and thus what constitute the life-history of a single organism 
get strung out into a series of conventional types. 

This inflated form does not persist long, breaking down gradually into : — 

Var. CLAViFORMis^ n.var. (Text-flg. 16.) 

Oellulae claviformes; corpore metabolico, anteriori latissimo, inflato, pone 
eonico acuminato; posteriori subcylindrico, fronte rotundato; paramylo nullo. 

Long. c. 40, lat. c. 14/x. With the type. 

Syn. Euglena pisciformis Dangeard {non Klebs) in These Proceedings, 1921, 
p. 121, PI. iv., f. 9-11. The shape of the cell is here club-shaped, simply inflated 
a little below the centre, acuminate behind and more or less conical in front. 
Much more common than the other forms. The chloroplast in all tends to be 
confined to the median portion of the cell, leaving the front and hind parts 
hyaline. It seems to be diffused or else composed of minute flecks. The nucleus 
is not observable and there are no paramylon rods or granules. 

A smaller form: — ^Long. 30-32, lat. 10-12/x is often found. 

xix.- — On Euglena pisciformis Klehs. (Text-fig. 17.) 

In note ii. (on the examination of certain green fluffy tufts, supra) I re- 
corded the occurrence of a form of Euglena pisciformis Klebs. Having just 
lately obtained Klebs' original figure and description it is evident that mine in 
Austr. Frw. Magell. is not correct. 

EuGLEKA PISCIFORMIS var. GRANULOSA^ n.var. 

Dififert a forma typica pyrenoidibus geruinatis, eellulae ad tertiam partem 
pone dispositis; parte anteriore magnis amylaceis nucleis digitt^tis repleta. 

Cell. long. c. 32, lat. Sj/.. Lismore. 

Compare Lemmermann, Eugleninae (in Pascher, op. cit., H. 5, p. 125, f. 
182 after Kllebs). The type is torpedo-shaped, rounded in front and drawn out 
by degrees evenly behind to a sharp point. The internal arrangements are very 
characteristic as the cell is furnished with 2 clear, transparent, homogeneous, 
parietal ehloroplasts which are very distinct and each contains a large pyrenoid, 
one fore and the other aft of the centre of the cell. The dimensions are given 



224 NOTES ON FRESHWATER ALGAE^ 

as 25-26^ long, 7-8jx broad. A smaller forra, var. minor Hansgirg is quoted at 
18-20/A long, 4^-5/* broad. 

Our specimens differ from the type in having the two pyrenoids side by side 
at the posterior third of the cell, while the portion anterior to them is crowded 
with distinct digitate amylaceous granules lying side by side at right angles to 
the cell- wall. The nucleus I did not see but the C.V. showed up more plainly 
than usual close to the stigma and pharynx. The vegetative cell is subglobose, 
diam. 16/x, and very often the characteristic chloroplasts and pyrenoids are 
visible in it. It is generally surrounded by a delicate irregular mucous or mem- 
branous investment. I have not observed the stages of gTOAvth. 

A type very close to Eil pisciformis Klebs is Eu. vivida Playf. 

XX. — Development of Euglena soeiabilis Dangeard. (Text-fig. 18.) 

Eu. sociahilis would seem to be a very rare species as Lemmermann says of 
it :- -'^Bislang nur aus Trankreieh." I was so fortunate as to obtain (May, June, 
1917 and even as late as October in the same year) a perfect series of its 
vegetative stages leading up to the motile form reproduced by me in Australian 
Frw. Flagellates (These Proceedings, 1921, p. 117, PL iii., f. 8, 9). The cell 
undergoes development within a wide transparent hyaline investment (schieim- 
huUe) often obviously stratified. The forms and dimensions are as follows: — 

a. — Cell spherical, diam. 25-38ju,; mucus diam. 95-126^. Cytoplasm finely 
granular, chlorophyll diffused, stigma generally distinct. Syn. Gloeocystis 
(Cklorococcum) infusionum Schrank. 

b. — Cell as in a but subglobose, long. 30, lat. 25^; mucus spherical, diam. 
120^. 

c. — Cell oblong, slightly pointed behind, with a small nick in front, long 53, 
lat. 36^; mucus spherical diam. 160^. Cytoplasm finely granular, chlorophyll 
diffused, stigma subapical. In these stages of development the characteristic 
digitate chloroplasts are not yet formed. 

d. — Cell oblong, somewhat attenuate posteriorly, acuminate behind, a small 
nick in front, sides very slightly retuse, long. 63, lat. 30^; mucus spherical, diam. 
160/A. Chloroplasts digitate, stigma subapical. 

e. — Cells subcylindrical, somewhat attenuate behind, subcapitate in front, 
sides lightly retuse, long. 74, lat. 21-27/i,, one to four in a mucous sphere diam. 
160-250yu,. Chloroplasts digitate, directed obliquely backwards, central lumen of 
the cell full of globular paramylon granules; pharynx expanded into a small 
nick in front; stigma distinct. (Text-fig. 18.) 

This series of vegetative forms leads up to Dangeard's zooid type, long. 85- 
95, lat. 21-28/x and it seems very probable that my undescribed specimen. I.e., PI- 
iii., f. 7, is really a more highly developed stage of this species. The disposition 
of the cytoplasm is the same, with the exception of the anterior chloroplasts 
which, as I did not observe tliem, are very likely disciform. Dimensions long. 
110, lat. 14/x. 

xxi. — Life-history of Euglena torta Stokes. (Text-figs. 19-23.) 

On February 11th, 1922, I gathered abundance of a floating Palmella stratum 
of Euglena vegetative cells in course of development. 

a. — Spherical resting cells, diam. 25/a, cytoplasm finely granular, chloroplasts 
irregular patches, stigma distinct. (Text-fig. 19.) 

h. — First motile form, the centre of one side of the resting cell is produced 



BY THE LATE G. I. PLAYFAIR. 



225 




Text-figs. 16-30. 

16. Euglena australica var. elaviformis, n.var. (x 800). 17. Euglena 
pisciformis, var. granulosa, n.var. (x 1200). 18. Euglena socidbilis (x 800). 
19-20. Euglena torta StoJces. (x 400). 21. Euglena torta var. ohesa, n.var. 
(x 600). 22. Euglena torta var. curta, n.var. (x 800). 23. Euglena torta 
Stokes. 24-25. Euglena geniculata (x 600) . 26. Euglena geniculata var. gut- 
tula, n.var. (x 600). 27. Euglena geniculata var. juvenlis, n.var. (x 600). 
28. Colacium ovale, n.sp. (x 1000). 29. Colacium arcuatum, n.sp. (x 1000). 
30. Exuviella lima (x 800) . 



226 NOTES ON" FRESHWATER ALGAE^ 

into a short conical prominence, the opposite side into a somewhat longer tri- 
angular one. (Text-fig. 20.) 

c. — Var, OBESA^ n.var. (Text-fig. 21.) 

Cellulae eorpore transverse conieo, uno latere rotundato, alioque retuso, inter- 
dum modice torto, superne eonico inferne cauda triangulari instructo. 

Second motile form, dimensions about 30 x 20^, body transversely conical, 
rounded one side, retuse the other, a conical prominence in front and a triangular 
tail behind, sometimes slightly twisted. 

d. — Var. CURTA^ n.var. (Text-fig. 22.) 

Cellulae breviores, magis oblique spirales, parte media constricta anteriore 
et posteriore inflatae, fronte in eonico, pone in cauda triangulari sine spina 
protractae. 

Third motile form, about 35 x 12yu,, constricted in the centre, inflated above 
and below, in front conical behind triangular, without a spine. A relatively large 
oblong nucleus often distinctly visible in the hinder part. 

e.—Euglena torta Stokes. (Text-fig. 23.) 

Compare Lemmermann, Eugleninae (in Pascher, op. cit., H. 5, p. 130, f. 
191). 

He quotes 63/x long, and the figure works out at 12yu, broad. Our specimens 
are shorter, about 40^ long and 14^ broad, cytoplasm finely granular in all 
forms; chlorophyll diffused or in a thin parietal lamina; tip and tail often 
hyaline; stigma always distinct; a large oblong nucleus in the hinder part; no 
paramylon rods or granules. 

From the plankton of a wayside waterhole on the Katikati Road, Tauranga, 
N.Z. 

xxii. — Life-history of Euglena genieulata Dujardin. 

In the early part of 1922 I obtained some scanty material from a road- 
side drainage pool. Later I treated this with hay infusion, some blowflies fell 
in and decayed and a culture of brown rot fungus resulted. It was in the 
examination of this fungoid growth that I discovered among the hyphae a whole 
series of forms illustrating the development of a very graceful little species of 
Euglena. After considerable hesitation I came to the conclusion that this should 
be identified with: — 

Euglena geniculata Dujardin. (Text-figs. 24-27.) 

Histoire Naturelle des Zoophytes Infusoires, 1841, PI. 5, f. 15, 16. 

The figTire given in Pascher, op. cit., Heft 5 (Eugleninae by E. Lemmermann, 
f. 206) does not in the least resemble this species. Dujardin's figures are strap- 
shaped or subeylindrical, rounded in front and acuminate behind, ending in a 
small prickle-shaped tail set obliquely. The flagellum is short and weak. 
Dimensions are quoted as 70-85yu, long and 12-22^ broad. Compare also Euglena 
mutabilis Sehmitz (I.e., p. 131, f. 203) which however is produced behind into a 
long spinous tail. 

With these essential details our specimens agree. The chloroplast appears 
to be a thin parietal lamina; nucleus median and globose with a cylinder of 
paramylon fore and aft of it. In the interspaces a few scattered granules. 
Stigma small but distinct. Membrane spirally striate obliquely from left to 
right. 

Cell. long. 100, lat. 12/,,. Tauranga, N.Z. 

Three successive non-motile vegetative forms: 



BY THE LATE G. I. PLAYFAIR. 227 

a. (Text-flg. 24). — Broadly ovate, adherent by the slightly pointed apex, long. 
21//,, lat. ITfi, develops forward and backward along the anterp-posterior axis. 

b.c. (Text-flg. 25). — Cell pointed and produced in front, sometimes also be- 
hind, long. 30^, lat. 16^. Internal details of all these and the succeeding forms 
are the same. 

d.e. — Var. guttula, n.var. (Text-fig. 26.) 

Cellulae guttulae formes vel clavatae, fronte acutae valde attenuatae; pone 
latissimae, rotundatae vel levissime acuminatae. Cytoplasma granuloso. Chloro- 
plastides laminae parietales tenues. Nucleus globosus in media cellula. Stigina 
parvo distincto. 

Cell. long. 34, lat. 14^. Cum prioribus tribus. Tauranga, N.Z. 

These two more developed forms noted both motile and non-motile. Cell 
drop-shaped or elavate, rapidly attenuated and sharp-pointed in front, broadest 
towards the rear where it is either broadly rounded or conical and acuminate. 
Cytoplasm minutely granular. Chloroplast a thin parietal lamina. Nucleus 
globular median. Stigma small but distinct. C.v. often visible. 

/. — Var. JUVENILIS, n.var. (Text-flg. 27.) 

Cellulae fusiformes, subeylindricae, fronte rapide attenuatae, acutae; pone 
eonicae acuminatae, interdum spinis minutis singulis instructae. Cetera, ut in 
var. guttuld. 

Cell. long. c. 50-60, lat. 10-12^. Cum priori. Tauranga, N.Z. 

A young form of the type; the apex still undeveloped, being much at- 
tenuated and acute; the nucleus lies a little below the centre; the terminal spine 
iwhieh in the type attains a length of 12/a, was only occasionally noted. All these 
forms were present in quantity. 

xxiii. — On tivo new forms of Colacium. 

Among other organisms observed in the Euglena-Shn that formed the subject 
of investigation in Note xviii. were Entomostraca bearing great numbers of 
Colacium. 

Colacium ovale, n.sp. (Text-fig. 28.) 

Cellulae primum ovales, deinde longe-ovatae. Zoosporae lineari-ellipticae, 
lateribus paene parallelis, fronte aeute-rotundatae ; pone levissime acuminatae, 
ChlorojDlastis in fragmentis inaequalibus dispositae. Stigmate nullo. 

Cell veg. long. 10-18, lat. 8/x. Zoosp. 24 x 6^. On a Copepod. Lismore. 

Cells at first oval, then long-ovate as in C. elongatum Playf. (These Pro- 
ceedings, 1921, p. 116, PL iii., f. 4-6). The zooid however was not the same 
shape, being linear-elliptic, acutely rounded in front and slightly acuminate be- 
hind. The chloroplasts were irregularly fragmented, and neither stigma, nucleus, 
C.V., nor pyrenoids were visible. 

Colacium arcuatum, n.sp. (Text-fig. 29.) 
Cellulae arcuatae uno latere quam levissime concavae paene planae, alter o 
latere modiee eonvexae; apicibus aeute-rotundatis. Cytoplasma minute granu- 
loso; ehloroplastidibus in fragmentis inaequalibus dispositis. Nee stigma visibile 
fuit, nee nucleus, neque c.v. 

Cell. long. 16-36, lat. 5-10^. On a species of Copepod. Lismore 
Only the vegetative cells were seen, attached as usual by a mucous pedicel 
singly or in clusters of 2-4. They were very slightly arched, being almost flat 
on one side and more or less convex on the other. Cytoplasm finely granular. 
Chloroplasts in irregular fragments. Neither pyrenoid, stig-ma, nucleus nor c.v. 
visible. 



228 NOTES ON FRESHWATER ALGAE. 

1 lost the specimens before I bad a chauce to detach any of the vegetative 
cells, otherwise a few minutes would have sufficed for them to develop into 
zoospores. It is noteworthy that though the cells become motile and produce 
flagella when detached, they show no more internal detail than was originally 
present. 

DINOFLAGELLATAE. 

xxiv. — On Exuviella lima (Ehr.) Schutt. (Text-fig. 30.) 

In 1909, when I gathered this organism alive from the creek at Canley Vale, 
my simple student's 1-6 in. obj. was not sufficiently powerful to show the necessary 
detail. On taking up work with better objectives in the Richmond River Dis- 
trict which is very rich in all kinds of flagellates, I had great hopes of coming 
across it again, but in vain. My notes therefore in "Peridineae of New South 
Wales" (These Proceedings 1919, pp. 817, 818") had to be written on preserved 
material. It is curious then that on starting work with the microscope in New 
Zealand this organism should be the first living thing to meet my eye, adding a 
new record to its distribution in freshwater. 

SJiape. — Our New Zealand specimens are oblong, rounded off at all points^ 
those from the suburbs of Sydney are generally somewhat narrowed towards the 
flagellate end. Subcircular specimens and truncate forms may also be noted, 
such being the result of transverse self -division. In side view the cell is much 
compressed and generally slightly arched. 

Cytoplasm. — The contents of the cell, and its disposition, are so obscured bj 
a layer of large irregular (paramylum ?) granules lying close together over the 
surface of the cell-wall, that it is difficult to see internal details. 

Cliromatophore.—Yellow-hTOwn or yellow-green in colour, apparently a de- 
licate parietal lamina. 

Nucleus. — A large globose or transversely oblong body occupying the pos- 
terior end of the cell and extending forward as far as the pyrenoid. It has the 
appearance of being punctate or finely granular as shown by Penard in Gym- 
nodinium mirahile and G. helveticum (Les Peridiniacees du Lac Leman). 

Flagella.— There are two flagella, one stout and up-standing, about the length 
of the cell. The other, corresponding to the transverse flagellum in Peridinium, 
is very delicate and wavy and lies close to the margin of the cell. A notch in 
the front valve, not found in the other, gives exit to both flagella. 

Pyrenoids. — ^A large pjrrenoid occupies the centre of each valve, being pressed 
elo.se against the cell-wall to the exclusion of the granules, hence showing up very 
clearly. From their appearance in side-view they seem to be either plano-convex, 
or disciform with slightly retuse centre. 

Contractile vacuole. — I have not been able to locate this, even with the 1-12 
in. obj. 

Stigma. — None, nor have 1 ever seen the red oil-drop sometimes present in 
Gymnodinium, etc. 

Habitat. — Noted as a marine organism from the Parramatta River and salt- 
water swamps at Newington (Sydney), both the ordinary size and var. major 
Playf., I.e., p. 818. In these marine specimens the membrane was quite smooth, 
not even punctate. 

Dimensions. — Freshwater, 25-36^, long, 20-26/m broad. Var. major 42-56ju 
long, 32-45 ij. broad. 

Suburbs of Sydney (N.S.W.); Tauranga (N.Z.). 



229 



STUDIES IN THE VEGETATION OF ARID AND SEMI- ARID 
NEW SOUTH WALES. 

Part i. The Plant Ecology of the Barrier District. 

By Marjorie I. Collins, B.Sc, F.L.S., Linnean Macleay Fellow of the 

Society in Botany. 

(Plates xv.-xxiii. and six Text-figures.) 

[Read 27th June, 1923.] 



Introduction. 

The State of New South Wales may be conveniently divided into three main 
physiographic units, the Coastal Plain, the Main Divide and the Great Western 
Plains. 

The Coastal Plain varies in width according to the nature of the rocks of the 
n^ain divide and their resistance to denvidation. Thus it narrows down at the 
MaePherson Range, the boundary between Queensland and New South Wales, 
where the hard volcanic rocks capping the Clarence Basin offer an effective re- 
sistance to denudation, while it widens out and reaches its greatest inland ex- 
tension of 60 miles (Jose, Taylor and Woolnough, 1911, p. 4) where the soft 
rocks of the coal measvires are encountered in the Hunter Valley near Newcastle. 

The formation of great estuaries and harbours along the coast of New 
South Wales, such as those of Jervis Bay, Port Hacking, Botany Bay, Port 
Jackson, Broken Bay, Newcastle, Port Stephens, etc., is thought to be due to the 
great earth movement which brought about the upward warping of an old 
Eastern Australian peneplain to form the main divide, and the simultaneous 
downward movement of the coastal strip, which necessarily resulted in the 
"drowning*" of river valleys and the production of deep harbours (Jose, Taylor 
and Woolnough, 1911, p. 6). 

The Main Divide occupies a broad belt of country roughly parallel with the 
coast and extending inland for a distance of from 150 to 200 miles. In it may 
be recognised a series of tablelands of altitudes^ ranging from a few hundred to 
over 6,000 feet, which mark the result of the uplift of the Eastralian pene- 
plain during Tertiary times. 

The faulting and warping which accompanied this uplift are responsible 
for the "blocking" of the plateau, the more or less flat tops of the hills, and for 
the presence of "rift valleys" or "geocols" breaking up the divide into separate 
tablelands. Where rivers leave the divide on its eastern and western flanks, ex- 
tensive dissection has taken place and plains of alluviation have been built up. 

The Great Western Plains extend from the western flanks of the Divide 
across the Darling River to the South Australian border. 




230 VEGETATION OF ARID AND SEMI-ARID NEW SOUTH WALES^ i., 

For the greater part they consist of old denuded peneplains, more or less 
buried in their own waste, and recent alluvials such as the black-soil plains of 
the Darling, Murray and Murrumbidgee Eivers and their tributaries. In no 
place is the altitude of these western plains greater than 1,000 ft. Isolated re- 
siduals of harder rocks are occasionally found standing out above the surround- 
ing country, but their height is accentuated by the extreone flatness which is a 
notable feature of the greater part of western New South Wales. 

The chief of these regions of high land in the western plains are the Cobar 
peneplain, 125 miles east of the River Darling, and the Barrier and Grey Ranges 
in the far west and north-west respectively. 

Not only are these high lands made up of some of the oldest rocks in the 
State, but they are famous for the important mineral belts which occur in them. 
These are the gold deposits of Mt. Brown near Milparinka, the silver-lead-zinc 
belts of Broken Hill in the Barrier, and the copper and gold deposits of the 
Cobar region. 

g -fOf/ru6 tO- fSf**^ /f- )LO«*«a X.O r dt-9 o*^ 



Sea, lev /i 



CodlHtssures 

Text-fig. 1. — ^Section across New South Wales from Barrier Range to 
Sydney (after David) . 

The red sandy soil which characterises the greater part of these western 
plains, and in which the old residuals stand partly buried, is held by geologists 
to be largely due to the wearing down of the extensive marine deposits which 
were laid down upon the floor of the vast Australian mediterranean sea in 
Cretaceous times. These marine sediments (Desert sandstones of the G-reat 
Artesian basin) extend from Queensland into north-western New South Wales, 
and across to the Lake Eyre region of South Australia. The presence of a por- 
tion of the Grreat Artesian basin in this region of New South Wales is of vast 
importance in a part of the country suitable only for scattered pastoral holdings. 

The south-western portion of the plains, the Riverina, is occupied by a large 
area of sediments laid down in the Eocene sea Avhieh invaded Southern Aus- 
tralia, even as far as Wentworth on the River Murray. In this belt of country 
sub-artesian water may be found. 

The interesting geological changes which took place in late Cretaceous and 
early Tertiary times in Australia,, and which went far towards giving Western 
New South Wales its present physiognomy, have been summarised by David 
(Jose, Taylor and Woolnough, 1911, p. 11) thus: — "In Cretaceous geological 
time we see rivers like the Bogan, Maequarie, Castlereagh, Namoi, Gwydir, etc., 
or rather ancestors of those rivers following their present general trend, empty- 
ing by separate mouths into the Cretaceous sea, whieli then washed against the 
western foothills of our Main Divide Vjetween Dubbo and Yetman. Later a 
gradual uplift of this old Cretaceous sea took place from the Gulf of Carpentaria 
to Coonamble (in N.S.W.), the hinge of the movement being perhaps along the 
Cobar to Broken Hill axis. This uplift joined by the method of "Engrafting" 
the Bogan, Maequarie, Castlereagh, etc., rivers on to tlie Darling River, The 
direction of flow of these rivers to the N.N.W. or N.W. is thus easily under- 



BY MAR J OKIE COLLINS. 231 

stood, as well as tl;e unusual angle which they make at their junction with the 
Darling River, a newer river than they, started on its career by the north to 
south tilt of the Cretaceous basin. They thus form, with the Darling, the 
gigantic "boat-hook bends" so well described by Mr. T. G. Taylor.* 

The North to South tilt of the Australian Continent in late Cretaceous time 
depressed the southern part of Australia beneath the ocean, so that the waters of 
the Eocene Sea crept inland to beyond Wentworth, and at least as far as the 
Arumpo bore. 

Later still a slight tilt took place in an opposite direction bringing the sedi- 
ments of the Eocene sea above sea level, engTafting the Murray, Murrumbidgee 
and Darling, and pushing the coast line from above Wentworth down to its 
present position on the Coorong." 

Since the receding of the Eocene sea from Southern Australia, western New 
South Wales has been exposed to the forces of denudation. At the same time 
its rivers have been responsible for extensive alluviation, and recent alluvials of 
the Darling and Murray and their tributaries are known to be underlain by great 
deposits of Tertiary and Post-Tertiary muds, sands and gravels. 

At the present day Western New South Wales is in the region of lowest 
rainfall in the State. By referring to a rainfall map for Australia (Text -fig. 
3) it will be seen that the isohyets are roughly parallel to the coast. When one 
passes from the Main Divide, where the rift valleys or geocols markedly affect 
the curves, and approaches the western plains, with their comparative uniformity 
and low relief, the isohyets are free from loops and become continuous curves. 

Owing perhaps to the fact that the moisture-laden south-easterly prevalent 
winds of New South Wales lose the gTeater part of their moisture in the region 
of the Main Divide, the western plains are characterised by decreasing rainfall 
as one proceeds west. The Barrier District falls for the most part within the 
10 inch isohyet, while Yandama on the western slopes of the Grey Range, is the 
region of lowest rainfall in the State, having an average annual total of 6.55 
inches. 

That the western portion of New South Wales is one of increasing ai'idity 
within recent geological time, is amply evidenced in the nature of the main and 
subsidiary stream channels. The river Darling itself, which traverses this western 
division of the State almost diagonally, shows such features of old age as ex- 
tensive meanders, oxbow bends and anabranches. Further indications of increas- 
ing aridity are to be found in the fact that tributaries of the Darling on its 
western side, such as the Warrego and the Paroo, now only reach the Darling 
after excessive floods in their upper coui^es. 

The Darling itself, being fed from higher regions where the rainfall is 
greater and more regular, is never entirely dry and only rarely ceases to flow. 

Thus the western plains of New South Wales present themselves to us as a 
vast area of country becoming progressively drier as we proceed west, a region 
which shows evidence of increasing aridity within recent geological time — a region 
where old residuals of ancient rocks are being relentlessly worn down to spread 
vast sheets of rubble and sand over the plains. 

Classification of Arid Regions. 
The Arid belt of Australia is of the typical flask shape, and, extending in 

*T. Griffith Taylor. The Physiography of the proposed Federal Territory at 
Canberra, p. 8, Figs. 6, 7. 



232 



VEGETATION OF ARID AND SEMI-ARID NEW SOUTH WALES,, i., 



an east-west direction from Broken Hill in the Barrier Range of New South 
Wales, to North West Cape in Western Australia, it occupies 37% of the area 
of the Continent. 




Text-fig. 2.— Western division of New South Wales showing isohyets and con- 
tours (after Griffith Taylor). 

It is where the eastern fringe of this arid belt abuts upon western and south- 
western New South Wales, that zones of varying degrees of aridity occur, corre- 
lated with the decrease in rainfall as one proceeds from east to west. 

Although there are many to whom the word "desert" is anathema when ap- 
plied to any portion of New South Wales, yet even the most hopeful must re- 
cognise, that there are often periods of years, when the greater part of the laod 
west and north-west of the Barrier Range presents a desert facies. 

What are the criteria for determining whether a country shall be classified 
as Desert, as Arid, or as Semi-arid? 



BY MARJORIE COLLINS. 



233 



The amount of rainfall in the year is universally recognised as being an 
important factor influencing plant covering in these parts. 

Yet while realising the vast importance of rainfall and the remarkably 
quick response to rain by plants in drier regions, the additional effects of tem- 
perature and evaporation should not be overlooked. 




Text-fig. 3. — Average annual rainfall map of Australia (after Griffith Taylor), 
Note small region in New South Wales with less than 8 inches rainfall. 

Various investigators have, at dif£,erent times, attempted to establish some 
basis of classification for regions of low rainfall. 

Warming recognises amongst arid floras the three zones, Desert, Shruh 
Steppe and Grass Steppe, but gives no quantitative climatic basis for this zona- 
tion. 

He describes arid regions as those "where rain falls at quite irregular times 
and is continuously lacking for a long time during the season otherwise most 
favorable to plant growth" (1909, p. 273). Amongst American workers, 
MacDougal establishes a basis for comparison by using extremes in rainfall 
amounts (1914, p. 175) while Forrest Shreve compares the rate of evaporation 
of the atmosphere with the moisture content of the soil (1915, p. 92). Dr. 
Cannon, in his recent investigation of the arid portions of South Australia, uses 



234 VEGETATION OE ARID AND SEMI-ARID NEW SOUTH WALES^ i., 

the average annual rainfall as a convenient method of establishing boundaries 
for botanical purposes. He refers (1921, p. 2) to a region having 5 inches or 
less of rain per annum as a desert, between 5 and 10 inches as arid, and be- 
tween 10 and 15 inches annually, as semi-arid. 

In establishing boundaries between desert, arid, and semi-arid zones, Koppen 
refers to the use of definite climatic factors which influence the distribution of 
vegetation. In reference to arid regions, he holds that an essential feature is the 
absence of run-off of water. Whei'e evaporation exceeds precipitation no streams 
originate. Since the temperature affects evaporation, Koppen (1918, pp. 193- 
203) holds that some combination of mean annual temperatui-e with rainfall 
should be used to fix the arid boundary. He gives the following table combining 
the effects of temperature and rainfall. 



Temperature 


77°F 


68°r 


59°F 


50°F 


41°F 


Desert Boundary . . 


13in. 


lliin. 


lOJin. 


gin. 


Sin. 


Steppe Boundary . . 


26in. 


23m. 


20Jin. 


18 in. 


16in. 



Since in the attempt to reach '^climax" formations nature must be con- 
tinually balancing and offsetting conditions of one year against another, it seems 
to the writer that the general physiognomy of the plant-covering (recognised as 
a direct response to the combination of the various factors — rainfall, temperature, 
evaporation, soil moisture, etc. — enumerated by the investigators mentioned 
above) should prove an effective means of determining the boundaries between 
desert, arid, and semi-arid zones. 

Ecologically we have come to recognise certain distinct types of plant for- 
mation in different parts of the world. 

For instance the Desert, a place where life at best is carried on under most 
adverse conditions, is of two types — the Ice Desert and the Sandy Desert. Again, 
transition zones are recognised from these desert types to less rigorous regions. 

These are the Tundras on the one hand, and the Steppes on the other. 

Developing upon these we recognise definite plant formations tending to- 
wards a ''climax'' wherever conditions are stable enough to allow a "climax" to 
be reached. 

However, whether a stable or an arrested climax is reached, the plant cover- 
ing over very wide areas is sufficiently uniform to give character to the region. 

Therefore it seems possible that from ecological studies, by reference to the 
botanical results, upon the country, of all the combined factors mentioned above, 
a classification of these zones should be more truly made. However, since more 
information concerning climax vegetation in the drier parts of New South Wales 
is necessary, it is the intention of the writer, for the purposes of the present 
paper, at least, to adopt the classification of arid regions used by Cannon and 
based upon average annual rainfall. ♦ 

Certain Features of Rainfall Affecting Arid Regions in Australia. 

In discussing rainfall one should not overlook the peculiar variations and 
disturbances which sometimes result in the "desert blossoming like a garden" after 
years of drought, or which, on the other hand, might despoil the semi-arid lands 
of their rich grassy herbage, leaving burning red sand between the trees and 
shrubs. 

There is no student of dry lands who is not familiar with the lavish response 
of the vegetation of these parts to good soaking rains. One cannot help but be 
impressed also with the fact that the vegetation of arid lands is influenced by 



BY MARJORIE COLLINS. 



235 



the time of the year in which the greater part of the rain falls, and perhaps 
even more by its reliability. 

From the economic point of view it would seem that the pastoral industry 
could be carried on successfully in regions of very low rainfall, provided the 
reliability of rainfall were good. 

In this connection Griffith Taylor has calculated the percentage of deviation 
from the average rainfall over a period of twenty years. This percentage is 
plotted (Taylor, 1918, p. 21) by isopleths on a rainfall reliability map (Text- 
fig. 4). 



« Thursday I. 




Lord Howe Is, 
Newcastle 



Text-fig. 4. — Rainfall reliability map of Australia with 10 inch isohyet marked 
(after Griffith Taylor). 



It is interesting to note that the 10 inch isohyet passes through the region of 
great reliability (15% to 20% deviation) in the south-west portion of the con- 
tinent, while in the Eiverina and in the Barrier District in New South Wales (the 
latter to be described in the present paper) the 10 inch isohyet passes through a 
zone of less reliability — of 20% to 30% deviation from the average. This fact 
is of great importance in connection with the extension of wheat-growing into 
arid and semi-arid lands. Taylor has given striking examples of the different 
types of rainfall in regions of the same average annual total. For instance, at 
Roeburne in Western Australia on the 15 inch isohyet .13 inch fell in 1891, 
while 42 inches were recorded in 1900. Cobar, in New South Wales, also on the 
15 inch isohyet, enjoys rainfall which falls almost uniformly throughout the 
year, whereas at Northam, in Western Australia, nearly all of the 15 ins. falls 



236 VEGETATION OF ARID AND SEMI-ARID NEW SOUTH WALES^ i., 

in the wheat-growing period — thus the latter is a farming region while the Cobar 
District is devoted to gTazing (Taylor, 1920, p. 332). 

Kainfall reliability is possibly a more important factor in the attainment of 
climax formations of the vegetation, than the actual rainfall total itself. 

In reference to the seasonal distribution of rainfall in Australia, Taylor re- 
cognises four rain regions : — Summer Rain Region in the North, Winter Rain 
Region in the South, Uniform in the East and Arid in the centre and Middle 
West. He maps Australia to show these rain regions and plots in an East and 
West line separating the regions influenced by Summer and Winter rain con- 
trols. Bj referring to the map, it will be seen that the central arid portion of 
Australia is affected occasionally by both these controls, and it will be shown 
later that, according to whether the summer or winter rain controls are dominant 
in the year, so will the nature of the ground flora be affected. 

The Winter rains are due to the westerly winds of Antarctic depressions, 
while the Summer rains are "largely due to the tropical cyclones which hover over 
Northern Australia when the sun is overhead in the Tropics, i.e., in Summer" 
(Taylor, 1920, p. 335). 

Another feature of rainfall, in relation to vegetation, is that emphasised by 
Dr. Cannon in his recent investigation on the flora of arid South Australia, 
namely, that of effectivity. 

A certain amount of recorded rainfall is known to be ineffective as far as 
the more permanent plant covering is concerned, since certain falls are sufficient 
only to reach the root-systems of the smaller herbage and annuals. Other falls 
again, often coming after long periods of intensely dry weather or drought, lose 
a considerable amount of their effectivity owing to the run-off from the hard- 
baked soil surface. The importance of successive falls of rain, particularly if 
these be slight, is urged. Cannon defines an ecologically effective rainfall as 
consisting of .15 inch or more, occurring in a rainy period (1921, p. 48). 

By comparing non-effective rainfall for various stations of decreasing rain- 
fall in South Australia, Cannon shows that the curve of aridity falls more rapidly 
than that of the recorded rainfall. 

Previous Investigations. 

Despite the great interest attaching to this region, both by reason of its 
vast extent (about 37% of the area of the continent) and of the great length of 
time during which it has been subjected to the forces of base-levelling and desic- 
cation, arid Australia has attracted comparatively few botanical investigators. 

What knowledge we have of its flora has been largely gleaned from the re- 
ports and collections of such Exploring expeditions as the Horn and Elder Ex- 
peditions, conducted from South Australia, and from the collections of private 
individuals. 

Only in 1918, when Dr. Cannon was sent to Australia by the Carnegie In- 
stitution of Washing-ton, was any attempt made to visit various stations in the 
desert and arid zones, with a view to studying typical plants in relation to their 
habitats and to one another. Although confined to the arid portions of South 
Australia, this investigation is of particular interest, since it is the first attempt, 
on a big scale, at undertaking a purely botanical research in the arid belt of 
Australia, and since the writer was able to bring to bear upon certain Australian 
problems his experience of desert floras in other parts of the world. 

However, since the observations of this writer were restricted to one season, 
and that a particularly dry one, it is apparent that a complete ecological investi- 



BY MARJORXE I. COLLINS. 237 

gation was impossible, and ttiat a certain percentage of the flora (chiefly annual) 
was not seen by him. 

In the preparation of vegetation maps for Australia certain geographers, 
such as Schimper and, more recently, Diels and Grrifflth Taylor, have mapped de- 
finite vegetational regions in the arid and semi-arid belts of Australia such as 
"spinifex and sand," "mulga," "mallee," "brigalow," "Savannah," etc. (Griffith 
Taylor, 1918, pp. 24, 26 and 27). 

It is interesting to note that, in Taylor's (1918, p. 27) vegetation map for 
Australia, based upon that of Diels, far western New South Wales, for some 
distance east of, and extending within the 10 inch isohyet even to the Lake 
Ejrre region in South Australia, is mapped entirely as Savannah. It is apparent 
that more detailed observation is needed in these regions, and it is the intention 
of the present writer to refer further to these vegetation maps in a later part of 
this series. 

Beyond references made by various geologists in their reports, the New 
South Wales portion of Arid Australia remains an untouched field for ecological 
investigation. 

Sir Douglas Mawson, in his report on the Geology of the Broken Hill Dis- 
trict, makes reference to the vegetation and recognises distinct habitats such as 
the hills areas where "mulga {Acacia aneura) scrub predominates," associated 
with which are Eremophila maculata, Fusanus acuminatus and others. He also 
notes (1912, pp. 230-231) the plains habitat with dominant salt bushes (species 
of Atriplex) and blue bush (Kochia). 

Mr. E. C. Andrews in a more recent memoir on the geology of the Broken 
HiU District (1922, pp. 33-36), also refers to the vegetation in a general manner, 
and draws attention to the extensive development of the Chenopodiaceae amongst 
the ground flora and the Myoporineae amongst the shrubs. Mr. Andrews' re- 
marks on the flora are accompanied by a series of interesting photographs of 
various species. 

Scope of the Present Investigation. 

Marking the eastern limit of the arid zone, this far western region of New 
South Wales should offer much that is of especial interest to the student of arid 
floras. 

It appeared to the writer that an interesting series of ecological studies could 
be made by visiting regions in each of the rainfall belts in Western New South 
Wales, correlating the vegetation, as far as possible, with rainfall and other 
climatic factors. 

Although the Broken Hill or Barrier District is not the region of lowest 
rainfall in the State, it comes within the 10 inch isohyet, and thus occurs near 
the eastern boundary of what is regarded botanically as arid country. It was 
thus considered by the writer as a suitable starting point for an investigation of 
this kind. 

In addition to the following work upon the flora of the Barrier District, the 
writer has investigated the flora of the Grey Range and its neighbourhood (a re- 
gion of lower rainfall, ranging from 6.55-7.5 ins.) and in addition the Cobar 
District, east of the River Darling — a region within the 15 inch isohyet. 

As the results of these expeditions amply justify the plan of the writer, it is 
intended to publish these as separate parts of a series. 

Apart from the purely scientific point of view, it is felt that in a country 
such as Australia, where a great portion of the wealth depends upon the pastoral 



^38 



VEGETATION OF ARID AND SEMI-ARID NEW SOUTH WALES^ i., 



industry and upon wheat-growing, any investigation which throws light upon con- 
ditions of life and growth in arid and semi-arid regions, should prove of great 
interest and value. 

The possibility or otherwise of extending our pastoral lands into the heart 
of the desert on the one hand, and that of subduing semi-arid lands for wheat- 
growing on the other, are questions continually before the Australian public. 

Only when we have accurate data, accurate climatic records, complete and 
searching investigations into the conditions of life and growth in these regions, 
shall we be able to judge without prejudice of the possibilities for opening up 
and settling this vast, waste interior of Australia. 

In connection with the present work the writer wishes to express her in- 
debtedness to Professor T. Griffith Taylor of the Department of Geography, Uni- 
versity of Sydney, who has kindly allowed her to use certain of his elimatologieal 
maps; to Mr. J. H. Maiden and his staff who have kindly given her access to the 
National Hei'barium and assisted in the identification of many species; to Dr. W. 
MacGillivray, of Broken Hill, whose wide knowledge of the Barrier District was 
always generously placed at her disposal; and to Mr, and Mrs. Morris of the 
Barrier Field Naturalists' Club, who have supplied much unpublished data, as 
well as material and general information. 

The writer also wishes to express her thanks to Sir Sidney Kidman, through 
whose hospitality she was accommodated and given transport on various sheep 
stations in the district, and to Mr. and Mrs. Foulis, of Corona Station, whose 
many kindnesses facilitated the investigation. 

The Barrier Range. 

The Barrier Range is made up of low hills of ancient contorted and meta- 
morphosed rocks which, with the exception of certain peaks, only reach a few 
hundred feet above the level of the surrounding plains. These hills which trend 
N.N.E. are regarded by geologists as being an extension of the Mount Lofty and 
Flinders Range systems of South Australia. They are continued further into 
Queensland by the Grey Range near Milparinka and Tibbooburra. 

Made up as they are of very ancient rocks, the Barrier Ranges have been 
subjected to denudation over a considerable period of time and give many evi- 
dences of peneplanation. Occasional high peaks, such as Mount Robe (950 feet 
above the level of the surrounding plains), Lewis Point, and Mount Euriowie, 
mark irregularities in an old peneplain surface estimated as being 1,400 feet above 
sea-level (Mawson, 1912, p. 218). 

The rocks making up this metamorphosed complex, are found by geologists 
to belong to two distinct and unconformable series, which have been named the 
Willyama or Broken Hill Series, and the Torrowangee Series. In his recent re- 
port on the Geology of the Broken Hill District, Andrews (1922, p. 107) has 
placed the age of the Willyama Series as Archaean and that of the TorroAvangee 
as Pre-Cambrian. 

The Willyama Series is made up for the most part of gneisses, schists, amphi- 
bolites, etc., amongst which occur the famous Broken Hill sulphide ore deposits. 
These rocks dip under the Torrowangee Series of quartzites, tillites, limestones 
and daystones at Poolamacca, 40 miles north of Broken Hill, while they outcrop 
again as an inlier 10 miles further north (Andrews, 1922, p. 38), where they ex- 
tend past Mt. Euriowie to Corona (Text-fig. 5). 

A feature of interest in the Torrowangee Series, and one which plays no 
small part in the establishment of vegetation, is the occurrence of enormous dykes 



BY MAEJORIE I. COLLINS. 



239 



CLAY PANS Ano 

SAND HILLS 
TO THE NORTH 



Corona 

CEucalyptus CiUei) 
Limestone 




Torrowar tj aee. Series 

fc/a ygtones, tjlliteslimestones.s/etes eg.) 

Intrusive GraruJbe 
Recent SanduPLcuns 



— EXTENDING 



D A RLIN a a I V t R 



Scale tn Afiles ~ 



ftJf^n*ngaon 



Text-fig, 5. — Geological sketch-map of the Broken Hill district, with plant associations 
indicated (based on map by Douglas Maiwson). 



240 VEGETATION OP AEID AND SEMI-ABID NEW SOUTH WALES, i., 

and irregular masses of white quartz, wliich break up and spread out over the 
plains and slopes as sheets of white siliceous rubble. The weathering out of 
quartzite boulders from the tillites of the Torrowangee Series is also responsible 
for the presence of rubble sheets on a larger scale. As will be shown later these 
rubble sheets form a particularly adverse habitat for the establishment of seed- 
lings. 

The plains by which the Barrier Range is surrounded extend on the eastern 
side for a distance of about 65 miles, falling gradually to the Darling River which 
is 600 feet below the level of Broken Hill. On the west they extend from the 
bold scarp characterising the western side of the Range, to thei Boolcoomatta 
Hills in South Australia. To the south the plains stretch beyond the Darling 
River through south-western New South Wales, into Victoria and South Aus- 
tralia, while in the north their extent is limited by the Grey Range and the Desert 
Sandstone Formation outcropping near Milparinka, and Tibbooburra. 

Andrews recognises three types of plain in the Barrier District — (1). The 
Mundi Mundi plain, a narrow alluvial plain bounded on the west by the Bool- 
coomatta Hills in South Australia, and extending in a northerly direction beyond 
Corona. 

(2). The Darling River type which he describes as "a very broad valley of 
alluvium dotted here and there with low rocky ridges and hills" representing "the 
gradual burial of an old system of valleys and inter- valley ridges" (Andrews, 
1922, p. 23). 

(3). The third type is the Northern Plain, in which only shallow patches of 
true alluvium occur, the riest being rolling stony and sandy country, marked here 
and there by stony residuals standing high above the general level. These northern 
plains are characterised by the extensive development of sand ridges 10 to 30 
feet in height, running approximately east and west, roughly parallel, and separated 
by hard clay-bottomed flats. 

Physiographic Features. 

The physiographic features of the Barrier Region are distinctly those of an 
arid cycle, in which transportation and redistribution of denuded material by 
wind are important factors. 

The low, undulating hills of the Barrier Range form but a slight relief from 
the monotony of the surrounding plains. Several creeks originate centrifugally 
in the hills, the chief of which are Stephen's Creek which flows in a south-easterly 
direction for about 125 miles to the Darling River plain, and the Umberumberka 
Creek which rises to the west of Broken Hill and loses itself eventually in a 
swampy region on the Mundi Mundi plain. 

The stream channels at the present time are little more than sandy courses 
lined by tall gum trees (Eucalyptm) and are known locally as "gum-creeks." 
The choking up of the channels with silt and sand and the cutting down of a 
narrower bed within the old stream channel indicate increasing aridity within 
recent times. Water rarely flows in any of these creeks at the present time, the 
rainfall being now insufficient and the evaporation too great. Local thunder- 
storms, however, often cause rapid flow of volumes of water, but only for a few 
hours. 

A further indication of aridity is to be found in the fact that the creeks 
generally lose themselves on the plains by spreading out into "lignum" swamps 
(Muehlenheekia Cunningharmi) . 



BY MARJORIE I. COLLINS. 



241 



Climate. 

Bainfall. — As has been shown above, the area under investigation is one of 
relatively low relief, hence there are no marked deviations in the rainfall curves 
for this portion of the State. The Barrier and Grey Ranges appear to have 
little influence on the nature of the rainfall, and this is also true of the Darling 
iliver basin, which is situated upon an old alluviated flood-plain. The average 
annual rainfall for the Barrier District ranges from about 8 to 10 inches. Corona, 
about 60 miles north of Broken Hill, has an average annual rainfall of 8.35 inches, 
while the average for Broken Hill, taken over 20 years, is 10.02 inches. 

A noteworthy feature of the rainfall of the Wiestem Division in New South 
Wales is its uniformity. 

The following graph, based upon that of Griffith Taylor (1918, p. 139) and 
the table of average monthly rainfall compiled from the records of the Weather 
Bureau, Sydney*, give some indication of the character of the rainfall and its 
uniformity in these parts (Text-flg. 6). 



INS. 



O^C. 



UAN 



FEB. 



MAR. 



4 PR 



MAY 



JUNl. 



Jl/L Atf6; SEP OCT. MOV. DEC 



8 coa/^a^2. 

C BROKEN ^<LL ^^ ' 




Text-fig. 6. — Graph representing average monthly rainfall for western towns in 
New South Wales (after Griflath Taylor). 

According to Taylor, such uniformity is due to the overlapping of summer 
and winter rain controls. In summer, with the southern movement of the sun 
accompanied by the chief wind systems, tropical and monsoonal rains affect the 
north of Australia, and even extend as far south as 32° Lat. In the winter 
(June), on the other hand, the effect of the Roaring Forties and Cyclones is felt 
as far north as Brisbane (Jose, Taylor and Woolnough, 1911, p. 203). 



*The weather records used in this paper were kindly supplied by Mr. Mares, 
Bureau of Meteorology, New South Wales. 



242 



VEGETATION OF AEID AND SEMI-ARID NEW SOUTH WALES,, 1., 



In the overlapping region rain results from both these sources, and a degree 
of uniformity follows. 



o<i 

cops 

I— I 

a 

-& 
o 

&(» 
I— i 

<! 



o 


oo(M-<*oocx)i>ooco 

Tj( N r-l T-l iH 




t-in t- to o 

OOiH -* t-CO 


CO o 

t-05 


C-J N rH 




> 
o 


00 <N N 00 iM 
00 00 rH CD t- 


t-00 

too 


iZ5 


CJ iH r-l 




o 


lO C3^ O O) rH 
05 LO iH 05 05 


coco 


O 


N tH r-l 




a 
© 


OS lO «0 to rH 
00 t> 05 C- 00 


-* 05 


CO 


(M H 






CO 05 iH rH TjH 
1-05 00 05 


CO 00 


CO rH 1-1 




"3 


00 CDC5 OrH 
00 CD 00 CD t- 





qothoo 

Tjl 00 f' 



•-5 ■* (M T-i 



in iH T-l rH rH 



-; 05COUt>OirHa:050(u 

a, CO t- o CD t- -i in lo ^ 



m rH tH 



> rH CO r-i cam 

> O 05 O -g t- 00 "5 

;j • ■ § • • fl 



rH 05 CO Tjl (M 


CCD-* 


a 


CO O -^O) t- 


O 05 [- 


o 


COCl rH 


g • • 


s 



-0-;,= — o C — 3 <^ 



ffl 



o 

fH 

Ph 
H 

O^ 

^^ 

MOQ 
«^ 

o'^ 
^^ 

>H 

> 



«^ 



as 






It will be seen from the graph (Text-fig. 6) that Cobar and Milparinka are 
not influenced by winter control, whereas Broken Hill in the Barrier Range 



BY MARJORIB I. COLJ^INS. 243 

shows a decided winter maximum. On Taylor's map of the rain control regions 
in Australia, the Barrier District occupies a position slightly to the south of the 
boundary between the summer and winter rain zones. 

With regard to rainfall reliability, the Barrier District comes within the 
legiou of medium reliability, the deviation being between 20% and 30% (Text- 
fig. 4). 

Although this deviation seems small in comparison with that of the desert 
region around Charlotte Waters, where the deviation is over 50%, the Barrier 
Region is nevertheless affected to a large extent by the periodic droughts — so con- 
dstent a feature of dry regions — and often may endure periods of more than a 
year when no effective rain falls. 

^ There is no doubt that these periodic rainless times exercise a controlling 
influence in giving the more permanent flora its distinctly xerophytic stamp. 

Temperature, Humidity and Evaporation. — The Barrier District resembles 
most arid regions in that its winters are cold and frosty and its summers very hot. 
The mean annual temperature for Broken Hill is about 64.5°, but during the 
summer there are often continuous periods of high temperature exceeding 100°. 
The mean maximum tempei-atures for January and February just exceed 92°, 
while extreme maxima for these months of 114.9° and 115.9° were recorded in 
1896. 

Since the relative humidity is lov/ (the wet bulb average not exceeding 62°) 
evaporation is necessarily high and in Broken Hill reaches an average of 85.2 
inches. The direct result of this high evaporation is the absence of actively 
running streams, except after heavy rainfall. 

Winds. — In arid regions, winds influence vegetation greatly in distribution, 
establishment and desiccation. 

The Broken Hill District is subjected periodically to violent windstorms, 
during which great quantities of dust and sand are transported. These dust 
storms usually occur in a dry season, when the ground is bare of its covering of 
undershrubs and herbs. 

Apart from these periodic visitations, there are the regularly occurring cir- 
cular whirlwinds known as "willy-willys" continually moving sand and debris. 
The building up of dunes and sand-ridges is taking place extensively in the 
Barrier region, on the Northern plains, and at certain places on the Darling 
River plain, at Menindie in particular. 

Wind-swept waste areas are commonly seen on pastoral holding's in the 
neighbourhood of Broken Hill, where over-stocking and over-gTazing took place in 
past years. 

These areas are particularly common on clayey regions where the wind has 
removed the surface soil, and the new clayey surface has been polished and 
hardened by the impact of wind-driven particles. These hardened bare areas 
resemble the clay-pans to be described later, and are particularly adverse habitats 
for the re-generation of vegetation, being unfavourable to the establishment of 
seedlings. 

Further evidence of the activity of wind in the Barrier District is to be 
found in the spreading out of great sheets of siliceous and ironstone rubble over 
the slopes and plains. 

Quartz and ironstone fragments found in these rubble sheets are the result 
of fragmental weathering of the large quartz and ironstone masses so prevalent 
in the range. The subsequent frosting of the quartz and intense polishing and 
rounding off of the ironstone, result from the impact of sand particles trans- 
ported by the wind. 



244 VEGETATION OF ARID AND SEMI-AEID NEW SOUTH WALES; i., 

Vegetation. 

As has been indicated above, the vegetation of the Barrier District is of a 
distinctly xerophytic stamp. 

The leaves are sclerophyllous and often hard and leathery in texture. 
Spinescence on a small scale is a common feature in the shrubs, as is also the 
development of prickles and other thickenings, characteristic of floras subjected 
to great drying. 

The development of clothing hairs is a noteworthy feature of numerous 
species of under-shrubs, such as the Amaranthaceae, certain Chenopodiaceae, 
Compositae, etc. 

Glandular hairs which secrete resin in abundance also occur frequently, and 
in such families as the Myoporineae and Sapindaceae are responsible for ^he 
lacquering of the leaves. Succulence is a feature which is developed very 
sparingly amongst the larger shrubs, but is commonly met with amongst such 
under-shrubs as the Chenopodiaceae, Zygophyllaceae, etc. 

In leaf-form there is a preponderance of the long narrow type such as in 
species of Eremophila, Heterodendron, Pittosporum and the phyllodia of several 
Acacias; these are either vertical or pendant, and assume the position, parallel to 
the incident rays of light, commonly found amongst xerophytic floras. 

Keduction of leaves and phyllodia is indicated in certain species, while 
aphylly occurs in such plants as Sarcostemma australe, Exocarpus aphylla, Tem- 
pletonia and others. 

Leaf -fall is not found in the time sense of the term, though certain species 
of Acacia and Cassia lose their phyllodia and pinnae in very dry weather, re- 
cuperation taking place after rains. 

In growth forma, the flora of the Barrier resembles that of other arid re- 
gions in that the Umbrella type of small tree or shrub is developed extensively. 
This is due to the peeuliarly compact manner of branching, — the laterals over- 
taking the main trunk and remaining close set, causing the typical canopy or 
umbreUa form. 

Amongst the annual ground flora the prostrate position is favoured and 
numerous species chiefly of Compositae adopt closely set mat forms. 

Ephemerals are well represented amongst the ground flora and in a bad 
season many annuals never reach more than dwarf form. 

In general configuration the flora of this region shows a preponderance of 
shrubs of varying sizes, amongst which species of Acacia, Cassia, Eremophila, 
Dodonaea, etc., are dominant. Small trees are, for the main part, but little 
higher than taU shrubs, species of Pittosporum, Hakea, Fxisanus (Quandong) 
and Myoporum (Sandalwood) being classified with these. 

Tall trees are relatively few, the most important being the River Red Gum 
{Eucalyptus rostrata), the Mallees {E. oleosa, ete.)i, the Beefwood {Grevillea 
striata), the Oak (Casuarina lepidophloia) and the Pine [Callitris rohusta) . 

The undergrowth is made up of small woody shrubs (Malvaceae, Solanaceae, 
Amaranthaceae, Chenopodiaceae) and under-shrubs (Chenopodiaceae, Zygophyl- 
laceae, etc.). 

Partly owing to the compact and non-spreading nature of the branching, 
and partly perhaps to the position assumed by the leaves, there is little or no 
shade cast by the shrubs and small trees in this district. However, as a result 
of the autumn and winter rains, there is a prodigal response of the ground flora, 
amongst which numerous families are represented. 

For the most part, Grasses are but sparingly developed in comparison with 
regions of greater rainfall and, when present, they generally adopt the tussock 
form. 



BY MABJORIE I. COLLINS. 245 

The Habitats. 

In the immediate neighbourhood of Broken Hill and for a radius of about 
twenty miles from the town the vegetation cannot be regarded as being at all 
representative. In the .early history of the development of the min ing township 
of Broken Hill, trees were ruthlessly cut down and used for all manner of pur- 
poses — mine timbering, homes for the miners, firewood, etc. With this wholesale 
destruction of tree and tall shrub vegetation, general deterioration of the flora 
naturally followed. 

With the coming of man's domestic animals, particularly the goat, the work 
of destruction continued, on account of the repeated browsing down of the 
seedlings of woody plants. 

At the present day the neighbourhood of Broken Hill presents a barren and 
desolate appearance, only a few trees and shrubs remaining on the higher slopes 
of the hills., the lower slopes and flat ground showing rocky and sandy patches 
with a sparse vegetation. 

Unnatural as these surroundings are, they are nevertheless interesting, since 
it is here that we can look .for signs of regeneration and succession; it is here 
also that the reeolonisation of waste lands may be studied, and the effect of an 
alien invasion noted. 

It is only when we pass beyond the limits of habitation, and beyond the 
town common, that the hills and plains give truer evidence of plant development 
and relationships. 

The fact that the district under examination is one of fairly low relief, 
accounts for the absence of sharply defined habitats such as are found in other 
parts of the world where arid regions extend to higher altitudes. 

If one makes allowance for a certain amount of overlapping on their boun- 
daries, one may recognise such habitats as rocky hills and slopes, creek beds and 
minor drainage channels, plains, and sandhills. The first bears a definite scrub 
flora of tall shrubs and small trees in which species of Acacia are dominant. 
There are also lower strata of smaller perennial shrubs and a ground flora of 
annuals, in which the Compositae are abundant. Where the slopes abut upon, 
or gradually pass into the plains on either side of the range, there is a mingUng 
of this ground flora with the Chenopodiaceae characteristic of the plain. 

The beds of creeks and drainage channels offer favourable conditions of 
establishment and, owing to the better water relations, large trees are found 
lining the watercourses, the chief being species of Eucalyptus (Gum) and Acacia. 
These better water relations apparently extend to the flats and slopes adjoining 
the creeks, where shrubs are found to attain a fuller growth. 

The plains habitat comprises areas of sandy soil, in which the sand some- 
times reaches a great depth, and in other places overlies a clayey sub-stratum. 
The chief plants populating the plains are perennial Chenopodiaceae, of which 
species of Atriplex and Kochia are dominant. 

There are certain distinct regions within the plains which may be regarded 
as minor habitats. These are the clay-pans and the flooded flats surrounding the 
larger of these. The clay-pans are bare, slightly depressed areas, roughly circular 
in outline, with hard clayey floors. 

In size they range from embryonic pans to extensive areas which form de- 
finite inland lakes after good .-ains. Interesting examples of large clay-pan lakes 
in the Barrier District are Lake Bancanya and Lake Cobham (Text-fig. 2). 

The clay-pans represent definite depressions in sandy country where water 
accumulates after rains. By evaporation, the salts contained in solution are de- 



246 VEGETATION OF ARID AND SEMI- ARID NEW SOUTH WALES; i., 

posited in the surface layers of the pans. Upon drying, the clayey floor of the 
pan becomes sun-baked and, after a dry season, extensive sun-cracks occur. The 
clay-pans are adverse habitats, chiefly on account of the hardness of the floor and 
the consequent difficulty of establishment of seedlings. 

Generally speaking small pans are bare but for a few Chenopodiaceae such 
as species of Babbagia and Bassia which are early colonisers of bare areas in 
these parts. A cane gTass {Glyceria ramigera F.v.M.) often completely reclaims 
clay-pans in the northern part, of the area. 

Large clay-pan lakes are generally bordered by flooded flats in which the 
dominant plant is the Black Box — Eucalyptus hicolor. 

These flooded flats or ''Box-Flats," as they are locally called,, apparently de- 
pend upon periodic inundation for their existence. It is in such flooded regions 
that the "Nardoo" (Marsilia spp.) forms close carpet-like communities. 

The Sandhills Habitat. — As in most regions where moving sand is a feature 
of the landscape, all stages may be observed from small transitory hummocks 
through embryonic and mobile dunes, to the more or less fixed sand-ridge, with 
its more stable vegetation. 

Since sand-ridge vegetation will be dealt with more fully in the second part 
of this series, the present paper refers only to the vegetation of the small mobile 
dunes often found in the neighbourhood of Broken Hill. 

Apparently the water relations are fairly good on these dunes since there is 
abundant covering vegetation — chiefly annual, after good rains. Salsola Kali is 
an early coloniser. Prostrate mat-growths are very common and act as good 
sand binders. There is very little tree vegetation developed on these smaller 
dunes — ^species of Casuarina and CalUtris appearing sparingly. 

The Vegetation of the Bocky Hills and Slopes. 

As has been shown, the rocky hills of the Barrier Region are made up, for 
the most part, of a complex of exceedingly old rocks of Archaean and Pre- 
Cambrian Age. 

The former of these, gneisses, schists, etc., yield upon denudation a coarse 
sandy soil, which soon becomes of the red colour characteristic of arid regions. 
The latter, the Torrowangee Series, are characterised by extensive deposits of 
slates and limestones, and it will be shown that these outcrops show some 
differences in plant covering. 

Owing to the absence of any appreciable altitude, there is not observed any 
marked difference between undulating summit and slope habitats. 

Where the rocky hills are intersected by creeks and minor drainage channels, 
slight differences in vegetation are noted, but only in the introduction of certain 
new species regarded as local invaders. Where the slopes merge into the plains, 
however, large sheets of rubble are spread out, and differences are more marked. 
These will be dealt with later in reference to rubble slopes. 

Mulga Scrub. — The vegetation of the rocky hills and slopes is comparatively 
luxuriant. It is distinctly a sclerophyllous scrub or bush, in which the genus 
Acacia is dominant. This scrub is known locally as Mulga Scrub, so called since 
the most characteristic species over large areas is the mulga, Acacia aneura 
F.v.M.* 

*The term mulga is sometimes used for other species of Acacia, as /J linnphylla 
the "sandhill mul^a" (Cannon 1921, p. 60). Acacia aneura is the species most*' 
widely known as Mulga in Australia and in this, as in other instances, indiscriminate 
use of popular names only leads to confusion. ^ 



BY MARJORIE I. COLLINS. 247 

A tall, fairly compact shrub; or at best a small tree, the mulga, in this dis- 
trict, reaches a height of from 10-15 feet. (PL xv., figs. 1, 3). 

The umbrella-form of growth adopted by so many perennials of arid regions 
is markedly developed in Acacia aneura and is a feature by which it is known 
throughout Australia. Forms are recognised also exhibiting differences in width 
and texture of the phyllodia. 

In certain parts of the district other Acacias assume the role of co-dominant 
with, or sub-dominant to Acacia aneura, the most important of which is Acacia 
tetragonophylla F.v.M. a large, compact, prickly shrub with minute phyllodia ar- 
ranged in gToups of two and three. From the impenetrable thickets sometimes 
formed by this species, as well as from its prickly phyllodia, it has come to be 
known as the "Dead-finish" (PI. xv., fig. 2). 

Other Acacias are A. cana J.H.M., A. Oswaldi F.v.M., A. Carnei J.H.M., 
A. sentis F.v.M., A. Burkittii F.v.M., A. Loderi J.H.M., etc. 

Second only to the Acacias in importance here are the Myoporineae, re- 
presented by many species of Eremopliila and more sparsely* by Myoporum and 
Pholidia. The Myoporineae are an eremian family par excellence and, though 
mostly shrubs in form, exhibit great variety in leaf type and protective device. 
The beautifully marked and coloured bell-shaped flowers of the Eremophilas are 
very striking in a bush where grey-gTeen is the prevailing tone. 

Together with the species of Dodonaea (Sapindaceae) found in the mulga 
scrub, some of the Eremophilas afford interesting examples of the phenomenon 
of leaf-lacquering due to the secretion of resin by glandular hairs. Other species 
of Eremophila exhibit, in varying degree, the development, of clothing hairs. It 
is the intention of the writer to refer, in a later paper, to the possible correlation 
of this development of clothing hairs and glandular hairs with the internal leaf 
structure. 

Although the greater number of plants in this habitat are tall shrubs of 
uniform size there are, nevertheless, certain trees entering into the association. 
The most important of these are Fusanus acuminatiis R.Br, (the Quandong), 
Casuarina lepidophloia F.v.M. (the "Belah" or "Oak"), Grevillea striata R.Br. 
(the "Beefwood") and Eucalyptus oleosa F.v.M. (the mallee). 

These trees are scattered as a rule, the Quandong and Beefwood being found 
freely mingled with the shrub vegetation. 

The "Belah" or "Oak" appears to seek better light relations, however, evi- 
dence of which is found in the occurrence of small "oak" communities on the 
higher ridges such as that near Thackaringa. 

The mallee is not common in the Barrier Range, being represented by a few 
isolated "pockets." On such high elevations as Mt. Robe, the mallee assumes 
more control and covers large areas on the hillsides. Other localities for 
Eucalyptus oleosa are the hillsides at Thackaringa and Apollyon Valley, where 
the mallee spreads from the floor of the valley to the summit of the hills. 

Other interesting trees of this association are Myoporum platycarpum R.Br. 
(the Sandalwood), Pittosporum phyllyraeoides D.C., Santalum lanceolatum R.Br., 
Exocarpus aphylla R.Br., and Hakea leucoptera R.Br, (the needlewood). 

The undergrowth in this habitat is for the most part made up of low shrubs 
and under-shrabs, and a ground flora of numerous types of annual plants. 

Most important of the low shrubs are species of Cassia, such as Cassia Sturtii 
R.Br., Cassia eremophila A. Cunn. and Cassia artemisioides Gaud. Others are 
Prostanthera striatiflora F.v.M., Senecio anethifolius A. Cunn., Sid-a virgata 
Hook., and Solatium Sturtianum F.v.M. 



248 VEGETATION OF ARID AND SEMI-ARID NEW SOUTH WALES, i., 

The under-shrubs chiefly occur on the lower slopes, or where degeneration 
has taken place owing to clearing, burnLng or other causes. They are mostly 
species belonging to the Chenopodiaceae more commonly met with on the plains 
habitat, such as species of Atriplex, Kochia, Enchylaena. Other Chenopodiaceae 
enter .extensively into the ground flora, but more particularly on the lower slopes 
where these merge into the plains. Various species of Bassia are common in 
such places. 

Certain grasses enter into the composition of the ground flora, species of 
Aristida and Stipa being noteworthy; other important grasses are species of 
Dantlionia, Pappophorum, Neurachne, etc. Since the tussock form is adopted in 
this region, the grasses very rarely colonise large areas to the exclusion of other 
species. 

The bunch grass or "spinifex,'' Triodia irritans R.Br., which occurs over 
such wide areas in the sandy desert of Central Australia, is not common in the 
Barrier Range. It has colonised a bare region beyond Stephen's Creek fairly 
extensively (Plate xxi.) but, apart from this locality, is only found in a few 
patches on the range where sand has accumulated. 

Parasites in the mulga scrub association belong to two families — the Loian- 
thaeeae and the Santalaeeae, the latter being root parasites. Of the Loran- 
thaceae, LorantJiiis Quandmig Lindl., L. exocarpi Behr., and L. miraculosus Miiq. 
are the most common species in the Barrier District and occur freely on many 
species throughout the scrub. The Acacias are most heavily affected and one 
commonly finds groups of trees killed by the parasite. Species of Acacia which 
have been weakened by the attack of Loranthus are commonly infected with the 
fungus TJromycladium Tepperianum, by which the work of destruction is com- 
pleted (PI. xviii., fig. 10). 

The Santalaeeae are small evergreen trees of the scrub and are represented 
by Exocarpvs aphylla R.Br., Santalum lanceolatum R.Br, and Fusanm acumin- 
atus R.Br, (the Quandong). 

On the slates and limestones of the Torrowangee Series, which outcrop in 
the hills north of Broken Hill, near Euriowie and Torrowangee, and are found 
on the Poolamacea and Corona runs, a less luxuriant vegetation occurs (Plate 
XV., fig. 3). This may be due to the less favourable water relations, or to the 
physical and chemical properties of the soil, or perhaps to the fact that the slate 
outcrops merge into the plains by low undulations, such as near Old Corona, and 
are extensively grazed over. In view of the occurrence of grassland upon slate 
outcrops, such as in the Mt. Lofty Ranges near Adelaide (Osborn, 1914, p. 114) 
one might regard the almost pure groupings of Stipa found upon slatey downs 
north of Broken Hill as attempts to form grassland within typical steppe country. 

In the neighbourhood of Euriowie and northwards through Corona, the 
Archaean gneisses and schists dip unconformably beneath the slates, limestones 
and claystones of the Pre-Cambrian System. 

In these parts interesting junctions may be seen of the two rock types, bear- 
ing fairly distinct floras. This is particularly marked near Corona homestead, 
where there is a junction between an extensive limestone outcrop and a hill of 
gneisses and schists. On the latter, the mulga scrub association is found, though 
the number of individuals is considerably less than in typical scrub association. 
This is no doubt due to grazing and the browsing down of seedlings of woody 
perennials by both sheep and rabbits. 

The limestone florn on tlio other liand is characterised by the development 
of a belt of a rare malice Encahiphis Gnhi .T.H.M. (PI. xvii., fig. 7). The pre- 



BY MARJORIE I. COLLINS. 249 

vious records of the occurrence of this mallee in Australia have been for the 
Flinders Range in South Australia, Burracoppin in Western Australia and for 
Euriowie and Mundi Mundi in the Broken Hill District. It apparently occurs 
chiefly on limestone outcrops. 

W. Gill says of it "It covers scores of acres on the poor hill and rise sides 
on travertine and chert country and looks like a great blue or silver blanket 
over the coimtry. 

Eucalyptvs oleosa is the only other mallee anywhere within hundreds of 
mUes" (Maiden, 1912, p. 177). 

In the Barrier Range, E. Gillei is restricted to small "pockets" as is true) 
of the other species of mallee in these parts. 

In association with E. Gillei are occasional migrants from the mulga scrub 
association of neighbouring rises, but for the most part there is no dense under- 
growth of smaller shrubs such as occurs in this latter association. There is a 
ground flora of numerous species of Kochia and Bassia such as K. se(Mfoli(i 
F.V.M., K. pyramidata Benth., K. brevifolia R.Br., K. villosa Lindl., Bassia dia- 
cantha F.v.M., Bassia lanicuspis F.v.M., B. paradoxa F.v.M. and others. 

Restricted to regions at the immediate base of E. Gillei, and growing in its 
shade are clumps of Zygophyllum fruticulosum D.C. which make bright green 
patches against the dull green of the Kochia. 

Botanical Features of Interest in the Mulga Scrub Association. — Despite its 
density, one of the most noteworthy features of the mulga scrub is the com- 
parative absence of shade. This may be accounted for in the fact that there are 
but few plants with wide spreading branches, the upwardly directed lateral 
branches helping to ensure the compact nature of the growth form so prevalent. 
The nature of the leaves or phyUodia, their reduced! form and their vertical lie in re- 
lation to light are also responsible for this absence of shade. Drought resistant, 
xerophytic forms make up the greater part of the ground flora on this account. 

Another important feature and one observed by the writer throughout both 
arid and semi-arid regions, is the localised grouping of numerous plants of one 
species into a community. 

It appears that any one species of the mulga scrub association may be thus 
localised and often occupy a considerable space of country. This feature has 
been observed by Dr. Cannon for the hillsides habitat in the Copley District of 
South Australia (Cannon, 1921, p. 74). The present writer is inclined to believe 
that this is by no means a local condition, but to be found universally in regions 
of low rainfall; nor is it a condition restricted to perennials, but occurs freely 
amongst annuals. 

Cannon (1921, p. 74) accounts for this grouping, in the uniformity of en- 
vironmental conditions over large areas. He observed such "mono-specific" 
communities of EremopMla Freelingii F.v.M., Pholidia scoparia R.Br., Zygophyl- 
lum crenatum F.v.M., Cassia Sturtii R.Br., and Kochia sedifolia F.v.M. To this 
the writer may add numerous examples met with frequently in widely separated 
localities, the most common being perhaps such additional species of Cassia as 
Cassia eremophila A. Cunn. and C. artemisioides Gaud., Seterodendron oleae- 
folium Desf., Acacia Burkittii F.v.M., A. cana J.H.M., and various Chenopo- 
diaceae. Since all the species met with thus may occur mingled freely in mixed 
communities, and in the general scrub association, the writer is of the opinion 
that there is a further factor than that of environmental uniformity causing this 
condition. 

Erratic rainfall, long periods of drought followed by or inten-upted by 



250 VEGETATION OF ARID AND SEMI-ARID NEfW SOUTH WALES; i., 

severe local thunderstorms, soaking rains which influence certain areas of country 
while leaving others quite dry, might possibly account for the extensive regions 
often colonised by one species. In a country where seed is known to lie dormant 
over many years, and where certain species may reappear in profusion after a; 
long interval, it is reasonable to suppose that rainfall and other factors have 
been at length suitable for extensive germination of seed of that particular 
species. 

The same reason may be held to account for the occurrence of Cannon's 
"monospecific communities," and one may believe that in some past year, condi- 
tions have been suitable for a profuse germination of seed of the particular 
species forming such a community. 

According to the rate of germination of a species in comparison with that 
of other species, and the percentage of survival of the seedlings, the ultimate 
configuration of the flora for the particular region in which these species occur, 
is predestined. 

A rate of germination and establishment about equal to that of other species 
reached by the rains, and a mixed community of the most general type may be 
the result. 

A rate of germination in excess of that of other species together with a 
greater survival percentage and one may reasonably expect a "monospecific'' 
com m unity. 

Once a monospecific community of this kind is well established there would 
be difficulties in the way of mixing the community by invasion, or by later ger- 
mination of seeds of other species. 

Vegetative propagation from horizontal roots, both in the case of perennials 
and annuals, may account, in part, for the extent of the areas occupied by 
"monospecific" communities. 

Isolated Species and Small Groupings. — In addition to the monospecific com- 
munities an interesting feature of this association is the occurrence in it of iso- 
lated individuals of certain species, or of small groups up to about a dozen in- 
dividuals. Cannon says of similar groups observed by him in South Australia 
(1921, p. 76) "Whether such small communities or isolated individuals represent 
recent introductions, or survivals from populous mixed communities, are matters 
of interest but this study throws no light on the problem." 

The occurrence of small groups of individuals, at varying intervals, is no 
doubt due to the method of vegetative propagation so common amongst perennials 
of arid regions, by forming new shoots from horizontal roots, a short distance 
beneath the surface. 

Various species of Acacia, e.g., A. Loderi J.H.M., HaJcea leucoptera R.Br., 
Heterodendron oleaefolmm Desf. afford instances of this. 

But the occurrence of certain small, quite isolated groups, or of solitary in- 
dividuals, which apparently play no part in the general configuration of the 
flora at the present time, needs some further explanation. 

The occasional occurrence of such species of Eucalypivs as E. odorata Behr. 
and Schlecht and E. transcontinentalis J.H.M. (mallees), as at Stephen's Creek 
in association with E. rostrata F.v.M., and again of E. odorata and E. intertexta R.. 
T. Baker at Apollyon Valley, as well as the "pockets" of Eucalyptiis oleosa in 
this latter locality, at Tbackaringa, and Mt. Robe, might be associated with a 
time when denudation had not reached its present stage, when rainfall was 
greater and the vegetation more luxuriant. These regions may be regarded as 
inliers of a more luxuriant vegetation which have persisted during the pauperisa- 
tion of the flora with increasing aridity. 



BY MARJORIE I. COLLI.VS. 



251 



The species Petalostylis labicheoides R.Br. (Caesalpineae), occurring on tlie 
lower slopes of Mt. Robe* and Eriostemon linearis A. Cunn., on the higher levels 
of Mt. Robe and Moorkai Hill, about 5 miles north of Stephen's Creek Reservoir, 
occur sparingij' in the district, and may be regarded also as possible survivors 
from less arid times. Another interesting species occurring in scattered gi-oups 
and isolated individuals throughout the district is CaUitris rohitsta R.Br. This: 
species attains its largest size in the Barrier District, in creek beds, in associa- 
tion with Eucalyptus rostrata, as at Pine Creek, 18 miles south of Broken Hill. 
It is fairly plentiful on rocky hillsides, also in a few localities north of Mt. 
Robe, and at Mootwingee. Since the pine has been cut extensively in the past 
for timber, it is rarely found covering large areas to-day. 

The occurrence of an isolated individual such as Flindersia maculosa F.v.M. 
(the "Leopard-wood") (PI. xviii., fig. 11), a solitary plant which has been seen 
in the Barrier Range in a valley between Purnamoota and The Paps, may be 
regarded as a chance invasion from the east, where Flindersia maculosa reaches 
a fine development in regions of greater rainfall. As far as the writer is aware, 
this species has not been recorded further west than the locality mentioned above; 
it occurs freely in a belt of country about 50 miles east of the Range, however, 
having been recorded from such stations as Kars, Langawirra and Yancannia. 

East of the River Darling, with greater rainfall, Flindersia maculosa becomes 
a more abundant component of the scrub. It appears to the writer that many 
solitary individuals such as this must have been carried over their climatic 
boundaries and may be regarded as peripheral migTants from their more typical 
habitats. 

The following list contains the chief species of the Mulga scrub association 
of the rocky hills and slopes habitat, f 



TREES. 



CaUitris rohusta R.Br. 
Casuarina lepidophloia F.v.M. 
Grevillea striata R.Br. 
Hakea leucoptera R.Br. 
Santalum lanceolatmn R.Br. 
Fusanus acuminatus R.Br. 
Exo car pics apihylla R.Br. 
Pittosporum pliyllyraeoides D.C. 
Acacia Oswaldi F.v.M. 
A. Loderi J.H.M. 



A. Burhittii F.v.M. 
A. aneura F.v.M. 
Flindersia maculosa F.v.M. 
Heterodendron oleaefolium Desf. 
Eucalyptus oleosa F.v.M. (mallee). 
E. transcontinentalis J.H.M. (mallee). 
E. odorata Behr. and Schlecht (mallee) 
E. Gillei J.H.M. (mallee). 
E. intertexta R. T. Baker. 
Myoporum platycarpum R.Br. 
Eremophila longi folia F.v.M. 



Acacia aneura F.v.M. 
A. tetragonophylla F.v.M. 
A. Carnei J.H.M. 
A. cana J.H.M. 
A. sentis F.v.M. 
A. rigens A. Cunn. 
.4. Burkittii F.v.M. 
Cassia eremophila A. Cunn. 
C. artemisioides Gaud. 
Templetonia egena Sweet. 



TALL SHRUBS. 

T. aculeata Benth. 

Dodonaea viseosa Jacq. 

D. viseosa var. spathulata Benth. 

D. attenuata A. Cunn. 
Heterodendron oleaefolium Desf. 
Jasminum lineare R.Br. 
Eremophila longifolia F.v.M. 

E. oppositifolia R.Br. 
E. alternifolia R.Br. 



*Mr. A. Morris has also collected this species at Mootwingee, towards White 
Cliffs, growing beneath Eucalyptus rostrata in a creek bed. 

f Since a census of plants of the Barrier District is being prepared by Mr. A. 
Morris, of Broken Hill, it is the intention of the present writer to list only the 
most typical species. 



252 



VEGETATION OF ARID AND SEMI-ARID NEW SOUTH WALES, i., 



Atriplex vesicarium Hew. 
Kochia pyramidata Benth. 
Koehia sedifolia F.v.M. 
Kochia villosa Lindl. 
Kochia ap press a. 
Kochia triptera Benth. 
K. triptera var. erioclada. 
Enchylaena tomentosa R.Br. 
Ptilotiis nohilis F.v.M. 
P. ohovatus F.v.M. 
Acacia continua Benth. 
Cassia Sturtii R.Br. 
Petalostylis labicheoides R.Br, 
Bodonaea lobulata F.v.M. 
Sida virgata Hook. 
Lavatera pleheia Sims. 
Pimelea simplex F.v.M. 



SMALLER SHRUBS. 

P. petrophila F.v.M. 
Sarcostemma australe R.Br. 
Marsdenia Leichhardtiana F.v.M. 
Solanum Sturtianum F.v.M. 
S. nigrum L. 
S. esuriale Lindl. 
S. ellipticum R.Br. 
8. ferocissimum Lindl. 
Prostanthera striatiflora F.v.M. 
Eremophila lati folia F.v.M. 
E. Duttoni F.v.M. 
E. maculata F.v.M. 
E. Sturtii R.Br. 
PhoUdia scoparia R.Br. 
Isotoma petraea F.v.M. 
Scaevola spinescens R.Br. 
Senecio anethifolius A. Cunn. 



GROUND FLORA. 



Cheilanthes tenuifolia (Burm.) Sw. 
Stipa scahra Lindl. 
S. trichophylla Benth. 
Aristida arenaria Gaud. 
A. calycina R.Br. 
Pappophorum nigricans R.Br. 
Neurachne MitchelKana Nees. 
Themeda Forskalii Hack. 
Triodia irritans R.Br. 
Danthoma pemcillata F.v.M. 
Bassia diacantha F.v.M. 



B. hicuspis F.v.M. 

B. paradoxa F.v.M. 

Bassia sclerolaenoides F.v.M. 

(limestones). 
B. divaricata F.v.M. 
Zygophyllum fruticulosum B.C. 
Z. ammophilum F.v.M. 
Z. apiculatum F.v.M. 
Z. iodocarpum F.v.M. 
Sida corrugata Lindl. 
Ahutilon spp. 



B. lanicuspis F.v.M. 
In addition to this are numerous annuals, chiefly of the Chenopodiaeeae, 
druciferae, Compositae and Leguminosae, which are more typical of the plains 
habitat and will be listed for that association. 

Vegetation of the Rubble Slopes and Plains. 

Where the rocky hills habitat merges into the plains by more or less gradual 
slopes, the ground surface is covered with a certain amount of talus. A note- 
worthy feature of these talus slopes is the occurrence, already referred to above, 
of areas of fragmental quartz and ironstone. 

These sheets of quartz and ironstone are continually being broken into 
smaller fragments and, as a result, areas covered by the white or black rubble 
often extend far out into the plains. Regions such as this are to be found near 
The Paps and Euriowie, and are largely developed on Poolamacca, Corona and 
Nundoora runs. These may be called, for convenience, the rubble plains, in 
distinction to the sandy plains to be described later. 

The rubble plains are of necessity limited in extent, since they are in process 
of formation, and are restricted to those regions where large masses of quartz 
and ironstone are formed. 

The fragments are of varying size and angular, the quartz on the one hand 
being frosted white by the action of wind and sand-blast, the ironstone being 
reddish black and highly polished by the same agency. 

A habitat such as this offers considerable difficulty of establishment for seed- 



BY KAEJOKIE COLLINS. 253 

lings. The intense glare from the glistening white quartz, and from the highly 
polished ironstone, must increase the temperature in which the seedlings estab- 
lish themsielves. As an indication of the extreme inhospitality of such an en- 
vironment, there is a very poor flora developed. Indeed for miles one might 
pass over undulating slopes and plains, colonised by but one species. This is 
Kocliia brachyptera F.v.M.*, a small, prostrate plant, which pushes its way be- 
tween the stony fragments. Its elongated, prostrate stems are red in colour, and 
bear small sage-gTeen leaves, densely clothed in silky hairs. The reddish colour 
of the stem veiled by the silky clothing hairs, possibly accounts for the mauve 
tints given to the slopes colonised by this plant. A notable instance of this 
occurs a little beyond Old Corona. The intense hairiness of the stem and leaves 
is possibly a necessary protection against intense insolation. 

Wiere wind has deposited layers of sand over these rubble slopes, seedlings 
of various annuals are not slow in making their appearance. These are chiefly 
annual saltbushes such as Atriplex holocarpum F.v.M., A. halimoides Lindl., A. 
angulatum Benth., A. campanulatum Benth,, Bassia divaricata F.v.M., B. para- 
doxa F.v.M., etc. J such Composites as species of Helipterum, Helichrysum, Cras- 
pedia, Calotus and others which are more typically of the sandy plains habitat. 

Should the seasons remain good, these sandy areas are made moderately 
stable by the establishment of certain perennials such as Atriplex vesicarium Hew., 
Kochia pyramidata Benth., Kochia triptera Benth., and others, and appear a,s 
small patches of vegetation scattered through the sheet of rubble. 

With the coming of a dry season, however, the adverseness of this habitat 
asserts itself, the invaders die out, and the rubble plain becomes once again a 
wind-swept waste. 

The Plains. 

Extending beyond the rubble plains for many miles, and completely sur- 
rounding the Range are the plains proper. These have been shown (pp. 229-30) 
to extend in a northerly direction as far as the Grey Range near the Queensland 
border, east towards the River Darling, and west across the border into South 
Australia. The plains have a surface covering of red, sandy soil, held by geolo- 
gists to be the results of denudation of the great Desert Sandstone masses, which 
at one time covered considerably greater areas than at the present daj''. These 
sands in places reach a great depth, but in other localities are underlain by a 
clayey subsoil at no great distance from the surface. In the immediate neigh- 
bourhood of the Barrier, the plains areas represent depressions between the spurs 
of the range, and on this account are not so sandy as the wide expanse of plains 
surrounding the range. From its most characteristic species in the Western 
District, Atriplex vesicarium, this habitat derives its name of "saltbush plain.'" 

Atriplex vesicarium is a low silverj^-grey perennial shrub dotted fairly uni- 
formly over the plains (Plate xx.). It is apparently able to withstand long- 
periods of dry weather, though occasionally one comes across large areas of dead 
plants of A. vesicarium, which have no doubt suffered in some prolonged drought. 
Although in the depression plains noted above there is possibly an accumulation 
of salts, yet Atriplex vesicarium is apparently not restricted to saline flats. 

The presence of a long tap-root, such as is known to exist in this species, 
might indicate that a water-level is tapped in some clayey substratum, so that a 
sandy habitat is not necessarily evidence of the non-halophytie nature of this 
plant. 

*This plant is described as Bassia brachyptera by Mr. Anderson in this Part 
of the Proceedings. 



254 VEGETATION OF ARID AND SEMI-ARID NEW SOUTH WALES, i., 

It would be a matter of interest and value in a pastoral country such as 
Australia, where thie perennial saltbushes are of such importance in grazing, to- 
establish the factors limiting the distribution of such a species as Atriplex vesi- 
cariutn. 

Certain questions present themselves in relation to this plant. Is it bound 
by climatic or edaphic ties? Is its presence as a dominant plant over such wide 
areas merely an indication of a primitive stage in a succession which rarely 
reaches a more advanced climax, or is its presence a direct response to such a 
soil factor as the presence of salts in the regions reached by its root system? 

■ The widespread occurrence of Atriplex vesicarium in a variety of situations, 
its dominance over many miles of both sandy and alluvial soils as well as in de- 
pressions which undoubtedly possess a certain percentage of salts, seems to sup- 
port the former view. 

Evidence has been brought forward to show that the nature of the rainfall — ■ 
whether influenced by summer or winter rain controls — affects the dominance of' 
the saltbush amongst the lower shrubs and ground vegetation in Australia. Taylor 
says (1920, p. 337) "The major portion of the remainder of the region of very 
erratic rainfall is covered largely by Mulga {Acacia sp.) [Acacia aneura] inter- 
spersed with open salt-bush areas in the south, and with open grasslands in the 
north. Latitude 23° seems to be the boundary between them at any rate in the 
central regions, and this recalls the seasonal rainfall control distinguishing the 
sage-brush from the grass-land regions in U.S.A." 

Thus it is seen that the distribution of open saltbush communities accom- 
panying mulga scrub depends upon winter rainfall control, whereas grassland 
develops in the northern section of the arid zone in response to summer rains. 
The Barrier District being in a region of winter maximum rainfall, and being a 
little to the south of the boundary between summer and winter rain controls, 
comes within the area of saltbush domination. Grasslands, in which the Mitchell 
Grass {Astrehla pectinata F.v.M.) Flinders Grass {Iseilima MitchelU Andr.) and 
Blue Grass {Andropogon sericeus R.Br.) are the most important pastoral grasses, 
ai'e known to exist beyond the Queensland border in a northerly direction from 
the Barrier Eange. In regions where there is an overlapping of the summer and 
winter rain controls, there is a mingling of grass with saltbush ground flora. 
This condition is common in the Grey Range in certain habitats and will be dealt 
with in a later paper. 

In the Barrier Range, occasional grasses are found with A. vesicarium on 
the saltbush plains, but these are often restricted to areas subjected to flooding,, 
and rarely play an important role. The small grass, ScMsfmis f asciculatus , 
which has appeared in the Barrier District within recent years, is an interesting 
coloniser of waste and eaten-out pastoral lands (PI. xvii., fig. 8). Forming- 
small closely pressed mats on the surface of the ground, it makes an apparent 
sward on otherwise bare areas. It seeds profusely, and apparently the seed 
germinates readily, since this plant responds by profuse germination to repeated 
falls of rain. 

Atriplex vesicarium often assumes complete control over large areas and 
forms a pure association. More often it is found in association with Kochia 
sedifolia F.v.M., Kochia pyramidata Benth., and Kochia aphylla R.Br. The two 
latt«r species of Kochia are larger shrubs and grow into big clumps. In places 
Kochia aphylla and Bhagodia spinescens R.Br, cover large areas to the exclusion 
of other species. The large rounded plants of the latter cover more surface than 
other species of saltbush and are often over 4 feet in height. At The Overflow,- 



BY MARJORIE I. COLLINS. 255 

Corona, the KocMa-Bhagodda community extends on to country which is flooded 
after heavy rains. In this locality, a close carpet-like community of Marsilia spp. 
covers the ground (PL xx., fig. 19). 

Occasional trees of the saltbush plains such as Myoporum platycarpum, 
species of Acacia, Grevillea striata, etc.; may be regarded as recent migrants 
from the mulga scrub association of the hills habitat, or, on the other hand, may 
represent residuals of a previous more luxuriant vegetation in which the plains 
climax vegetation resembled that of the hills. 

Anmial Flora of the Plains. 

A striking feature of the plains is the profuse development between the 
perennial saltbushes of an annual ground flora in response to rains. These an- 
nuals are particularly noticeable on regions which have been heavily grazed over, 
and in these localities they may dominate the plains for many miles. 

The chief families represented amongst this ground flora are the Chenopo- 
diaeeae (annual saltbushes) Zygophyllaceae, Compositae, Leguminosae, Malvaceae, 
etc. Many other families are represented by but a few species. The annual 
saltbushes such as Atriplex holocarpum F.v.M., A. halimoichs Lindl., A. angu-j 
latum Benth., A. campanulatum Benth., A. limhatum Benth., A. Muelleri Benth., 
etc., are the first species to colonise minor drainage channels, and the edges of 
roadways and stock routes after rains. These species extend further on to bare 
land and prepare the way for the later development of Compositae and Legu- 
minosae a,nd other families. The dominance of annual Chenopodiaceae and Com- 
positae amongst the ground flora of the plains can no doubt be accounted for in 
the efficient means of seed dispersal adopted by these families and the consequent 
increased capacity for invasion. This capacity for invasion is still further in- 
creased owing to the greater fertility known to exist in families which produce 
only one seed in each flower (Clements, 1916, p. 65). 

Amongst the Compositae, species of Helipterum,, Helichrysiim, Craspedia, 
Myrioceplialus and many others make a brilliant display of colour. Almost pure 
communities are sometimes found extending over many acres. In growth form, 
this family exhibits great variety, species of Helipterum, Helichrysum, Butidosis, 
Millotia, Isoetopsis, Angianthus and others being found in ephemeral or dwarf 
form; closely-set mat forms in which the capitula are clothed in dense woolly 
hairs are found in such genera as Chtonocephalus, while rosette forms are favoured 
by Leptorrhynchus, Craspedia, etc. 

Amongst the Leguminosae pure communities of the mauve and purple Swain- 
sona and the scarlet pea (Sturt's Desert Pea) Clianthvs Dampieri A. Cunn. often 
replace the Compositae (Plate, xix.). 

Clianthus Dampieri is a prostrate plant, the grey-gTeen leaves and branches 
of which are clothed in long silky hairs. After a rainy season acres of plains^ 
country are ablaze with the large scarlet and black flowers of this species. Apart 
from the Chenopodiaceae (species of Bassia, Bahbagia, Arthrocnemum and 
Kochia) succulence is notably developed in the Portulacaeeae among-st the flora 
of the plains. Portulaca oleracea L., a small prostrate annual, and Calandrinia 
halonensis Lindl., the bright pink-flowered "parakeelya" with its rosette of rigid 
succulent leaves, are useful fodder plants on account of their water storage 
capacity. Sheep have been known to live on the latter plant for weeks without 
water. 

Another succulent species amongst the annual ground flora and a good fodder 
plant is Tetragonia expansa an important coloniser of bare, over-grazed regions. 



256 



VEGETATION OF ARID AND SEMI-ARID NEW SOUTH WALES^ 1., 



The following is a list of the chief species of the plains flora. The plants 
occurring in the clay-pans, crabholes and flooded flats are included in a later list. 



Gramineae* 

Stipa scabra Lindl. 

S. elegantissima Labill. 

Paiypopliorum nigricans. 

Danthonia penicillata F.v.M. 

Eragrostis falcata Gaud. 

Panicum, melananthum F.v.M. (rare). 

Aristida ramosa R.Br. 

Trirapliis pungens R.Br, (rare) . 

Trir aphis mollis var. humilis R.Br. 

Andropogon annulatus Forsk. (rare). 

Schismus fasciculatus. 

Hordetim inurinum (introduced). 

Avena fatiia L. (introduced). 

Astrehla pectinata F.v.M. 
Liliaceae. 

Tliysanotus sp. ^ <>• 1 -t 

BuJbine semibarbata Haw. f * " "^ 
Chenopodiaeeae. 

Bliagodia spinescens R.Br. 

Chenopodium nitrariacium F.v.M. 

Atriplex vesicarium Hew. 

A. holocarpum F.v.M. 

A. halimoides Lindl. 

A. angulatum Benth. 

A. Muelleri Benth. 

A. campanulatum Benth. 

A. fissivalve F.v.M. 

Kochia sedifolia F.v.M. 

K. ap)hylla R.Br. 

K. triptera Benth. 

K. triptera var. erioclada Benth. 

K. pyramidata Benth. 

K. brevifolia R.Br. 

K. ap press a Benth. 

K. eriantlia F.v.M. 



hrachyptera F.v.M, 
Georgii Diels. 



K. 
K. 

Bassia sclerolaenoides F.v.M. 
B. diacantha F.v.M. 
B. bieuspis F.v.M. 
B. quinqueeuspis F.v.M. 
B. lanicuspis F.v.M. 
B. longicuspis F.v.M. 
]B. obliquimispis Anderson. 
B. biflora F.v.M. 
B. f)icornis r.v.M. 
B. divaricata F.v.M. 
B. ventricosa J.M.B. 
B. limbata J.M.B. 



\B. intricata Anderson. 

Babbagia acroptera F.v.M. 

B. dipterocarpa F.v.M. 

Enchylaena tomentosa R.Br. 

Arthrocnemum sp. 

Salsola Kali L. 

Salsola Kali var. strobiUfera Benth. 
Amaranthaeeae. 

Ptilotus nobilis F.v.M. 

P. obovatiis F.v.M. 

Alternanthera nodiflora R.Br. 
Aizoaeeae. 

Tetragonia expansa Murray. 
Portulaeaeeae. 

Portulaca oleracea L. 

Calandrinia balonensis Lindl. 
Crueiferae. 

Stenopetalum lineare R.Br. 

Lepidium papillosum F.v.M. 

L. fasciculatum. 

Blennodia lasiocarpa F.v.M. 

B. trisecta Benth. 
Crassulaceae. 

Tillaea recurva Hook. 
Leguminosae. 

Swainsona procumbens F.v.M. 

S. tephrotricha F.v.M. 

S. phaeifolia F.v.M. 

Lotus australis Andrews. 

Psoralea patens Lindl. 

Clianthus Dampieri A. Cunn. 
Geraniaeeae. 

Erodium cygnorum Nees. 
Oxalidaeeae. 

Oxalis corniculata L. 
Zj^gophyllaceae. 

Zygophylhim fruticulosum D.C. 

Z. ammophilum F.v.M. 

Z. iodocarpum F.v.M. 

Z. apiculatum F.v.M. 

Nitraria Sclioeberi Linn. 
Euphorbiaceae. 

Euphorbia Drummondii Boiss. 

E. australis Boiss. 

E. eremophila A. Cunn. 
Malvaceae. 

Sida eorrugata Lindl. 

Lavatera plebeia Sims. 

Abutilon spp. 



*The Gramineae were named by Miss E. Reed, B.Sc, Department of Botany, 
Perth. W.A., to whom the v/riter's thanks are due. 

tThese .species are described by Mr Anderson in the present Part of the 
Proceedings. 



BY MARJORIE I. COLLINS. 



257 



Frankeniaeeae. 

Frankenia sp. 
Tbymeleaeea;e, 

Pimelea simplex F.v.M. 
Umbelli, ?era e. 

Didiscus glaucifolius F.v.M. 

Dauciis hrachiatus Sieber. 
Convolvulaeeae. 

Convolvulus eruhescens Sims. 
Boraginaeeae. 

Heliotr opium curassavicum L. 

Lappula concava F.v.M. 

Rochelia m,accoya F.v.M. 
Solanaceae. 

Solanum esuriale Lindl. 

8. ellipticum R.Br. 

S. ferocissimum Lindl. 

Nicotiana suaveolens Lehm. 
Goodeniaeeae. 

Goodenia sp. 
Cueurbitaceae. 

Gucumis sp. 
Compositae. 

Clear ia sp. 



Minuria sp. 

Galotis hispidula F.v.M. 

G. erinacea Steetz. 

Brachyeone sp. 

Isoetopsis graminifolia Ture. 

( ephemeral ). 
Myriocephalus Stuartri Benth. 
Vittadinia australis A. Rich. 
Angianthus pusillus Bentb. 
Gnephosis cyatliopappa Bentb. 
G. shirrophora Eenth. 
Eutidosis helichrysoides D.C- 
Leptorrhynchus squamatus Less. 
Helipterum strictum, Bentb. 
II. pygmaeum Bentb. (epbemeral), 
H. hyalospermum, F.v.M. 
H. florihundum D.C. 
II. m.oschatum Bentb. 
H. corymbiflorum Seblecbt. 
Helichrysum sp. 

Caloceplialus platycephalus Bentb. 
Craspedia globosa Bentb. 
Chtmiocephulus sp. 



Clay-Pans, Crab-Holes and Flooded Flats. 

Witbin tbe plains may be recognised certain minor babitats sucb as tbe clay- 
pans, and tbe smaller circular depressions known locally as "crab-boles" or "gil- 
gais." 

Tbe clay-pans represent shallow depressions in tbe sandy plain, from wbieb 
tbe sandy surface soil bas been removed by wind, leaving exposed a bard clayey 
floor. Tbe pans vary in size from embryonic pans to large areas some miles in 
diameter. Twin pans occur side by side and, upon tbe removal of tbe interven- 
ing barrier of sand, tbese mei-ge into one pan. It is possible tbat tbe largest pans 
have been formed by tbe merging of neighbouring pans. Being tbe centres of 
localised drainage systems within the plains, rain-water collects in tbe pans and, 
owing to the hard clayey floor, is unable to percolate away. In medium-sized 
pans tbe water may remain for a couple of months, while in tbe largest pans 
which are regarded as inland lakes, tbe water may remain for over a year be- 
fore complete evaporation takes place. Clay-pan country is thus of importance 
in a pastoral land where water conservation is one of tbe chief problems. With 
evaporation, tbe salts collected by the rain-water as it drains off tbe higher land 
are deposited on the floor of the pans, together with fine layers of silt. The 
clayey floor subsequently becomes sunbaked and hardened to such an extent that 
it offers an effective resistance to invasion. 

In dry seasons tbe pans, for tbe most part, remain bare and windswept, but 
after a light fall of rain such Chenopodiaeeae as Babbagia acroptera F.v.M., and 
Bassia spp. commence colonisation. These two genera represent the pioneer 
colonisers amongst the Chenopodiaeeae on bard clayey soils. With colonisation, 
even though scattered, a certain amount of sand accumulates in tbe pan around 
tbe pioneer colonisers, and this paves tbe way for a more extensive invasion. 
Small pans may pass through a complete cycle in this manner, until completely 
filled in with sand; they are then popvilated by tbe typical species of tbe plains. 



258 VEGETATION OF ARID AND SEMI-ARID NEW SOUTH WALES, 1., 

The large pans are rarely empty long enough for a complete colonisation to take 
place. During the recent drought in the Barrier District, however, a large clay- 
pan lake 100 miles north of Broken Hill, Lake Bancannia (PI. xxiii., fig. 25) be- 
came completely dry, and colonisation by species of Bassia and Chenopodium 
took place. A mixed community of species of Bassia, in which B. divaricata is 
dominant, has taken possession of the central portion of the pan, while large 
clumps of Chenopodium nitrariaceum F.v.M. mingled with Bassia spp. inhabit 
the outer zones. 

The larger clay-pans are often delimited from the surrounding country by a 
special flood zone which is likely to be inundated after heavy falls of rain, owing 
to the overflowing of the clay-pan lakes. These are known as flooded flats or 
"Bos" flats from the characteristic tree, the Black Box (Eucalyptus bicolor A. 
Cunn.), which forms a distinct belt around the clay-pan (PL xxii., fig. 23; xxiii., 
fig. 24). Eucalyptus hicolor is restricted to these flooded fla,ts in the Bamer District 
and apparently depends for its existence upon periodic inundation. At Lake 
Bancannia and Lake Cobham (PL xxiii.) the Box Flats had not been inundated 
for over two years at the time of the writer's visit in October, 1922, and the 
trees were consequently dying out. 

Another species commonly found on these flooded flats and occupying the 
ground between the Box trees is MueMenheckia Cunninghamii (Polygonaeeae) 
the "lignum." MueMenheckia Cunninghamii is a leafless shrub with long inter- 
twining branches. In places it forms impenetrable thickets, and may reach a 
height of over six feet. It is a species associated with flooded and swampy 
country throughout the western district of New South Wales. 

A ground flora develops upon these flooded zones in which species of Mar- 
silia, the "Nardoo," (PL xx., flg. 19) often form pure communities, spreading 
over many acres. Marsilia Drummondii and the diminutive Marsilia exarata A. 
Braun are commonly found in the Barrier District. Colonising wet soil either 
with Marsilia or forming pure communities is the small Composite, Centipeda 
thespidioides F.v.M, Mimulus repens R.Br, forms flat mats on drier soil, while 
Morgania glabra R.Br., Lavatera pleheia, Lappula concava F.v.M., Linum mar- 
ginale A. Cunn., Wahlenhergia sp. and Compositae such as Helipterum florihun- 
dum D.C., H. corymbiflorum Schlecht, Angianthus pusillus Benth. and many 
others occupy the outer fringes of the flooded areas. 

The crab-holes or "gilgais" are small circular depressions in which water 
lodges after rain. Certain investigators hold that these depressions are hollowed 
out by the "willy wiUy" winds (Pittman, p. 465) while others believe that they 
are due to surface subsidence, following the consolidation of alluvials below 
(Jose, Taylor and Woolnough, 1911, p. 122). The crab-holes are colonised by 
plants of tlie flooded flats habitat just described. Being small localised areas, 
they may pass through a complete successional cycle, commencing with wet soil 
forms such as Marsilia and Centipeda, and ultimately reaching a final stage of 
Composites and grasses, merging into the plains flora. Occasionally a final stage 
in the colonisation of the crab-holes is characterised by a bright green clover 
which stands out in striking contrast to the prevailing grey-green of the sur- 
rounding plains flora. At other times, the "gilgai pea," Swainsona procumhens, 
makes pleasing patches of colour in the crab-holes, with its masses of deep purple 
flowers. The following is a list of the chief plants of the clay-pans, crab-holes 
and flooded flats. 



BY MARJORIE COLLINS. 



259 



T^iliees. 

Marsilia Drummondii. 

Marsilia exarata A. Braun. 
*Gramineae. 

Eragrostis imhecilla Benth. (rare). 

E. lacunaria F.v.M. 

Deyeuxia quadriseta Benth. 

Bromus arenarius Labill. 

Hordeum murinum Linn. 

Chloris ventricosa var. tenuis R.Br. 
Amaryllidaceae. 

Crinum flaccidum Herbert. 
Polygonaceae. 

Muehlenbeckia Cutminghamii F.v.M. 
Chenopodiaeeae. 

Atriplex limhatum Bentli. 

A. eampanulatum Benth. 

A. angulatum Benth. 

A. holocarpum F.v.M. 

A. halimoides Lindl. 
Bassia divaricata F.v.M. 

B. lanicuspis F.v.M. 
Bahhagia dipterocarpa F.v.M. 
B. acroptera F.v.M. 
Chenopodium nitrariaceum F.v.M. 
Encl%ylaena tomentosa R.Br. 

Aizoaeeae. 

Tetragonia expansa Murray. 
Papaveraeeae. 

Glaucium flavum (introduced). 
Linaceae. 

Linum marginale A. Cunn. 



Cruciferae. 

Blennodia spp. 
Leg'uminosae. 

Psoralea patens Lindl. 

Swainsona procumhens F.v.M. 
Zygophyllaeeae. 

Nitraria Schoeheri Linn. 

Zygophyllum sp. 
Malvaceae. 

Lavatera plebeia Sims. 
Mj'rtaceae. 

Eucalyptus bicolor A. Cunn. 
Campanulaeeae. 

Wahlenbergia gracilis D.C. 
Borag'inaceae. 

Lappula coneava F.v.M. 

Rochelia maccoya F.v.M. 
Solanaceae. 

Nicotiana suaveolens Lehm. 
Scrophulariaeeae. 

Mimulus repens R.Br. 

Morgania glabra R.Br. 
Compositae. 

Centipeda tliespidioides F.v.M. 

Angianthus pusilliis Benth. 

Helipterum floribundum D.C. 

H, corymbiflorum Schleeht. 

H. stipitatum F.v.M. 

Cotula coronopifolia L. 



The Vegetation of the Creeks. 

The general monotony of the plains is relieved by the lines of trees marking 
the meandering courses of the creeks. 

Eucalyptus rostrata F.v.M., the River Red Gum (PL xxii., fig. 22), is the 
characteristic tree of these creeks and is a species apparently well suited to drier 
regions, since it is found in similar situations throughout arid Australia. Al- 
though the creeks in these localities are nothing more than sandy channels, in 
which vv^ater runs only for a short time after heavy falls of rain, the water re- 
lations are nevertheless good, since this is the only habitat bearing large trees. 

Cannon has shown that the main portion of the root-system of Eucalyptus 
rostrata is "made up of large horizontal members which may extend for a long 
distance from the central stem. Thus the superficial roots of Eucalyptus were 
seen to reach out nine meters or more and to lie at a depth not exceeding 60 em." 
(Cannon, 1921, p. 62). 

Although the development of a superficial root-system may be regarded as a 
response to the characteristic rainfall of arid regions, yet it seems probable, in 
the case of Eucalyptus rostrata at least, that some subterranean water level is 
tapped by the main vertical root. This seems the more likely when we take into 
consideration the lengthy periods of drought, when the surface layers of the soil 
are hardly moistened at all. 



260 VEGETATIOK^ OF AHID AND SEMI-ARID NEW SOUTH WALES;, i., 

In addition to Eucalyptus rostrata, other trees such as Pittosporum phylly- 
raeoides, Fiisanus acuminatus and more frequently species of Acacia, such as A. 
sentis and A. Burkittii, inhabit the creeks. Certain tall shrubs of the hills habitat 
border the creeks occasionally, particularly where these leave the hills a,nd enter 
upon the plains. 

Heterodendron oleaefoUum reaches a fine tree growth in such situations. 

The sandy flats which extend back from the creeks to the plains, and are oc- 
casionally subjected to flooding, are generally colonised by species of Atriplex, 
Kocliia and Ehagodia and eventually merge into the plains habitat. In certain 
localities a white alkaline scum is deposited on the surface of the soil after the 
falling back of flood waters. This alkali has been found to contain magnesium, 
sodium and lime sulphate, chloride and carbonate (Mawson, 1912, p. 223). 

In such places the "Sodabush," Nitraria Schoeheri, occurs abundantly, form- 
ing large rounded bushes. With Nitraria, species of Zygophyllum are often 
found. 

Minor drainage channels are bordered by pioneer Chenopodiaceae, annual 
species of Atriplex predominating. These are followed by perennials such as 
Atriplex vesicarium, KocMa pyramidata, Kochia aphylla and Bhagodia spinescens. 
A shi'ubb}" Composite — Senecio magnificus, the broom-like Templetonia egena 
and the annual Psoralea patens are common in such situations. 

On a small creek on Corona at the northern limit of the Range, the "paper- 
bark tea tree," Melaleuca parviflora LindL, was found in association with a few 
3'oung trees of Eucalyptus rostrata. 

This is the only locality in the Barrier District where this species was ob- 
served by the writer. 

As a general rule the first tree growth bordering young creeks is of various 
species of Acacia, Acacia sentis and Acacia Burkittii being important pioneers. 
An interesting plant of the sandy creek beds and rocky ledges bordering the 
creeks is the orange-flowered Amaryllid, Crinum flaccidum Herb. This species 
is one of the few geophytes of the district, and perennates by means of a bulb 
sometimes placed at a distance of over a foot beneath the soil surface. Within a 
week after a fall of rain, the red sandy soil of the creek beds and flooded zones 
is turned to a vivid green by the young leaves of Crinum bursting through the 
&oil. This plant forms large seeds which are unprotected by a coat other than a 
thin layer of cork and, not having any special means of dispersal, germinate 
almost immediately. It is quite usual to flnd numerous germinating seeds of 
Crinum lying on the surface of the sand. 

The cotyledon remains attached to the fleshy seed by means of a sucker which 
transfers food from the seed to the young plant (Rendle, 1904, p. 305). At 
quite an early stage the base of the cotyledonary sheath becomes thickened to form 
the first bulb scale and eventually, by elongation of the sheathing portion of the 
cotyledon, the plumule and the young bulb are carried beneath the surface of the 
soil. In this manner Crinum overcomes tlie disability of a naked seed by form- 
ing a special organ of perennation, a bulb, during the germination period, and 
still further by burying the bulb thus formed. 

Vegetation of the Sandhills. 

As has been shown above, there are large areas in the Barrier District 
covered with drifting sand. This sand is of the same nature as that of the 
pandy plains, and is thought to have been derived by the wearing down of the 
Desert Sandstones of Cretaceous Age. 



BY MAKJORIE COLLINS. 261 

Stages may be observed in the formation of sandhills, from small, transitory 
hummocks, through mobile small dunes, to the more stable fixed sand-ridge, with 
its climax vegetation. The latter is developed extensively in the northern part 
of the region and extends across to the Lake Eyre Basin in South Australia. 

As this sand-ridge climax will be dealt with more fully in the second part of 
this series, it is intended here only briefly to describe the flora of the small, mobile 
dunes in the immediate neighbourhood of the Range. 

Since these small dunes are unstable in formation and plant covering, and 
since their flora merges appreciably into that of the plains, they might be re- 
garded as a minor habitat within the plains. 

The pioneer colonisers on the-se small sandhills are Salsola Kali and its 
variety S. Kali var. strohilifera. Salsola Kali is the "roly-poly" of the arid re- 
gions of Australia and is well suited to dispersal by wind, the whole plant being 
snapped oft: at the base and tumbled over the plains. An interesting evidence of 
the dispersal of this species is to be found at the border fence between South 
Australia and New South Wales, where old plants of Salsola are piled high 
against the wire netting. Salsola Kali appears to favour the lower slopes of the 
dune during the early stages of its colonisation, but later advances up the slopes 
and occupies considerable areas. 

For the main part, the flora of these small sandhills is of the annual type 
developed upon the sandy plains. Being localised areas often of small extent, 
they are sometimes colonised exclusiveh/ by one species. Myriocephalus Stuartii 
and other Compositae sometimes completely invade young dunes. 

The occurrence of species not previously met on the plains, such as the 
shrubby Acacia ligulata and the broom-like Crotalaria dissitiflora, indicates some 
betterment of the water relations in this habitat. 

Amongst the annual species, the Compositae, Leguminosae, Malvaceae, Sola- 
naceae, Portulaeaceae and Gramineae axe well represented, and resemble closely 
the annual flora of the plains. Occasional new species are met, however, which 
are apparently limited to sandhills. These are notably the rigid Aizoon quaclri- 
fidum F.v.M. with its small succulent water-storing leaves, and the large prostrate 
hairy Psoralea eriantha Benth. which forms spreading mat-like colonies. 

That these dunes are still in a developmental stage is indicated by the almost 
complete absence from them of any tall shrubby or tree vegetation and the pre- 
ponderance of annuals amongst the ground flora. Later, the dunes become larger 
and more stable and a climax vegetation is reached similar to that described for 
the hills habitat, in which species of Acacia, Eremophila and Casuarina play an 
important role. 

General Discussion. 

As has been shown above, the vegetation of the Barrier Range is of two 
main tj'pes — that of the rocky hills and slopes, and that of the plains. The 
vegetation of the former habitat is mulga scrub, in which Acacia aneura is 
dominant and is associated with other species of Acacia, Eremopkila, Dodonaea, 
Sakea, etc. Occasional communities of Casuarina lepidopliloia (oak). Eucalyp- 
tus oleosa (mallee) and Callitris rohusta (pine) occur within this association. 

The plant covering of the plains is for the most part of open communities 
of Atriplex vesicarium (the saltbush), with which Kochia sedifolia (bluebush), 
Kochia apliylla (cottonbush), K. pyramidata and Rhagodia spinescens are often 
associated. Any of these species may assume the role of co-dominant with 
Atriplex vesicarium. 



262 VEGETATION OF ARID AND SEMI- ARID NEW SOUTH WALES-, i.^ 

An annual flora made up chiefly of Compositae, Cruciferae, Papilionaeeae 
and others appears on the plains after good rains. Within the plains association, 
minor habitats may be recognised, such as rubble slopes and plains, clay-pans, 
g'ilgais, gum-creeks, flooded flats and small mobile dunes. 

The rubble slopes and small dunes represent primary bare areas within the 
plains and, being but sparingly colonised and in the latter case primarily by 
annuals, they often revert to their original bar© condition with the coming of a 
drought. 

The clay-pans, gilgais and flooded flats are subjected to changing conditions 
according to whether the season is rainy or dry, and consequently rarely reach a 
stabilised condition of vegetation. The gum-creeks on the other hand represent 
better water relations and bear a vegetation of tall trees {Eucalyptus rostrata) 
and shrubs {Acacia spp.). 

The Barrier District has been largely devoted to the pastoral industry during 
the past 50 to 75 j^ears and this, together with the fact that the rabbit has become 
an established pest, must be considered as contributing, in part, to the present 
day configuration of the vegetation. The combined effect of sheep and rabbits 
helps to check the natural development of vegetation. This must have occurred 
to a great extent in the early days of the grazing industry when overstocking of 
runs was the rule. Overstocking, particularly when followed by drought, causes 
large secondary bare areas, from which the natural vegetation has been eaten out. 
In a 'continued drought, these areas become windswept wastes, the surface soil is 
often removed, leaving exposed a hard clayey substratum which becomes com- 
pacted and hardened by windswept particles. Such areas are particularly in- 
hospitable to the invasion and establishment of seedlings and thus regeneration 
of vegetation is retarded. The extensive burrowing of rabbits, the browsing down 
of seedlings and the ring-barking of woody perennials near the burrows are also 
responsible for secondary bare areas in scrub or plains. 

Since the saltbush {Atriplex vesicarium) and the mulga {Acacia aneura) 
are amongst the chief fodder plants of the district and represent the mainstay of 
stock during long droughts, it seems possible that grazing may have played a 
large part in preventing the development of the vegetation beyond the mulga 
scrub association on the one hand, and the saltbush association on the other. The 
browsing down of seedlings, the ring-barking of young woody perennials and the 
wholesale eating of seed in tinies of drought by stock and rabbits must also play 
some part in arresting the development of vegetation in these parts. 

As far as our present knowledge goes. Acacia scrub is the most extensively 
developed and the most characteristic vegetation of the hilly regions ih. the Barrier 
District, while the saltbush association is that of the plains. Further evidence 
should be obtained, however, from widely separated localities, before we can fairly 
judge whether these represent true climax associations for the particular climatic 
region involved. The occurrence of small groupings or communities of tree vege- 
tation, in the scrub association, such as that of the mallees {Eucalyptus oleosa) at 
Mt. Robe, and E. odorata and E. transcontlnentaMs at Apollyon Valley, and of 
the pine, Callitris robusta (Mootwingee) and the oak, Casuarina lepidopMoia 
(Thackaringa) points to the possibility of a potential climax more advanced than 
that existing over the greater area of the Barrier Range to-day. Either these 
communities are to be regarded as representing attempts at the further develop- 
ment of the vegetation from a mulga scrub pre-climax, or they represent residuals 
of a more luxuriant vegetation associated in former times with less arid condi- 
tions. Since the region under investigation is one of increasing aridity within 
recent geological times, the latter view appears the more i^robable. 



BY MARJORIE COLLINS. 263 

Evidences of sueoession in the Barrier District are but few. Where second- 
ary bare areas occur in mulga scrub owing to clearing, burning, and dying out of 
communities during droughts, these are colonised by Chenopodiaceae such as 
Atriplex vesicariwm, Kochia sedifolia, K. pyramidata, Enchylaena tomentosa, etc. 
In rare cases earlier stages of colonisation of secondary bare areas are observable, 
species of Bassia always appearing before the larger species of Atriplex and 
Kocliia. At various localities, and particularly as seen by the writer at Corona 
on rocky hills a pure association of Bassia makes up the ground flora., the chief 
species being Bassia divarieata, B. lanieuspis , B. hicuspis and B. diacantha with 
B. paradoxa on the lower slopes (Plate xv.). 

The Chenopodiaceae are apparently the pioneer colonisers on all bare ground 
in the Barrier District, whether of primary or secondary origin. The rubble 
slopes and plains described above are often colonised exclusively by one species, 
Kochia hrachyptera, and rarely reach a more advanced stage of development. 
Bare clay-pans and clayey flats subject to flooding are colonised in the pioneer 
stages by species of Bassia and Babhagia, followed later, in one instance at least, 
by Ghenopodium nitrariaceum (PL xxiii., fig. 25). 

Early colonisation of dunes takes place by Salsola Kali, while all drainage 
channels and minor watercourses are first invaded by annual species of Atriplex. 

On the limestone slopes of the Torrowangee Series, where only scattered 
members of the mulga scrub are represented, species of Kochia are pioneer 
colonisers, Kochia sedifolia being the dominant species in the open community 
which results (PL xvii., fig. 9). Low undulating slate outcrops are also colonised 
by open communities of Kochia (PL xvii., fig. 8). 

Prom these facts it would appear probable that the saltbush association of 
the plains represents an early stage in succession, prevented from further develop- 
ment by biotic as well as climatic factors. The occurrence of members of the 
saltbush association upon secondary bare areas within the mulga scrub association 
points to the developmental relationship of these two associations. 

Of the various tjrpes of vegetation investigated by Dr. Cannon in South 
Australia, that of Copley in the Flinders Range (1921, p. 64) resembles most 
closely that of the Barrier Range in New South "Wales. This is to be expected, 
since not only is the Barrier Range regarded by geologists as being of similar age 
and as representing an easterly extension of the Flinders Range, but climatically 
the regions are alike. Copley is shown to have an average annual rainfall of 8.40 
inches, while Broken Hill the chief town of the Barrier Range has 10 inches. 
The Copley District again resembles the Barrier in its position with regard to 
the summer and winter rain controls and in the variability of its rainfall. 

The vegetation of the Copley District resembles that of the Barrier in that 
there is a distinct scrub on rocky hills, in which species of Goissia, Eremophila 
and Acacia are important. The vegetation of the plains is made up chiefly of 
Chenopodiaceae with species of Atriplex and Kochia dominant. Cannon refers 
to the two genera Gassia and Eremophila as being probably the most typical 
genera near Copley, and refers to the association of the hills as the Gassia-Eremo- 
phila community (1921, p. 73). This appears to be the main point of difference 
between the scrub of Copley and that of the Barrier Range where species of 
Acacia are dominant. 

In order to establish truly the relationships of the vegetation of the Barrier 
District with those of other parts 0;f arid Australia, and to gain a truer concep- 
tion of the successional development and the nature of climax vegetation in this 
region, it would be necessary to enclose and make stock and rabbit proof a large 
area of hills and plains country. Until this is carried out on a sufaciently* large 



264 VEGETATION OF ARID AND SEMI-ARID NEW SOUTH WALES,, i., 

scale, it will not be possible to judge fairly of the natural developm-ent of this 
vegetation, whether progressive or retrograde, uor of its future possibilities as 
an economic asset. 

Summary. 

1. Phj^siographic units of New South Wales are described with especial re- 
ference to the nature and formation of the Grreat Western Plains. 

2. The classification of arid regions is briefly discussed, and the suggestion made 
that ^'climax'' vegetation should be used as a basis of classification. 

3. Certain features of rainfall affecting the arid regions of Australia are dealt 
with and an account given of the geologic, physiographie and climatic fea- 
tures of the_ Barrier Range. 

4. The plant associations of various habitats in the Barrier District are dis- 
cussed and lists of the most representative species given. 

5. The most important of these are the Mulga Scrub association of the rocky 
hills and slopes and the Saltbush association of the plains. 

6. Minor habitats such as rubble slopes, clay-pans and crab-holes, fiooded flats, 
g-um-creeks and sandhills are described, and an account given of their plant 
covering. 

7. In a general discussion the developmental aspect of the vegetation of the 
Barrier District is dealt with and comparisons made with the arid flora of 
other parts of Australia. 

Literature Cited. 
Andrews, E. C, 1922. — The Geology of the Broken Hill District. Mem. Geol. 

Surv. N.S.W., No. 8. 
Cannon, W. A., 1921. — Plant Habits and Habitats in the arid portion of South 

Australia. Carn. Inst. Wasli., Pub. No. 308. 
Clements, P. E., 1916. — Plant Succession. An analysis of the Development of 

Vegetation. Carn. Inst. Wash., Pub. No. 242. 
Jose, Taylor and Woolnough, 1911. — New South Wales, Historical, Physio- 
graphical and Economic. 
KOPPEN, 1918. — Klassifikation der Klimate nach Temperatur, Niederschlag, und 

Jahreslauf. Petermann's Mitteilungen. Vol. 64. (Review, 

1922 Monthly Weather Review, U.S.A., Vol. 50, No. 2). 
MacDougal, D. T. and Collaborators, 1914. — The Salton Sea. A Study of the 

geography, the geology, the floristics and the ecology of a desert 

basin. Carn. Inst. Wash., Pub. No. 193. 
Maiden, J. H., 1912. — A Critical Revision of the genus Eucalyptus. Vol. 2., Pt. 

5. 
Mawson, D., 1912. — Geological Investigations in the Broken Hill Area. Mem. 

Boy. Soc. S. Aust., Vol. ii., Pt. 4. 
OSBORN, T. G. B., 1914. — Notes on the flora around Adelaide. New Phyt., Vol. 

13. 
PiTTMAN, E. F. — Mineral Resources of N.S.W. 

Rendle, a. B., 1904.— The Classification of Flowering Plants. Vol. 1. 
SCHIMPER, 1903. — Plant Geography upon a physiological basis. 
Shreve, Forrest, 1914, — Rainfall as a determinant of soil moisture. Plant 

World, Vol. 17. 
Taylor, T. (}.. 1918. — The Australian Environment. Comvionweallh Adv. Coun. 

Sei. and Indnjitry, Mem. No. 1. 
, 1920. — Agricultural Climatology of Australia. Quart. Journ. 

Roy. Meteor. S'oe., Vol. 46. No.' 196. 
Warming, E., 1909.— Oecology of Plants. 



BY MARJORIE I. COLLINS. 265 

EXPLANATION OF PLATES XV.-XXIII. 

The writer has to thank Miss E. Reed, B.Sc, University of Western Australia 
for the photographs reproduced in figs. 3, 8, 9, 10 and 19;' Miss D. Nobes, B.Sc, 
for those in figs. 5, 7 and 18; and Mr. A. Morris, of Broken Hill for that in 'fig. 11. 

Plate XV. 

Fig. 1. — Group of mulgas (Acacia aneura) on rocky hill at Corona, north of 
Broken Hill. Ground vegetation made up of species of Bassia. 

Fig. 2. — Acacia ietragonophylla ("Dead-finish"). Ground flora — species of 
Bassia. 

Fig. 3. — Low hills near Corona with open mulga scrub. Corona creek with 
Eucalyptus rostrata in middle distance. 

Plate xvi. 

Fig. 4 — Sarcostemma australe ('"snake-busih" or "caustic-bush"), a leafless 
Acsclepiad growing on rocky slopes. Broken Hill. 

Fig. .5. — Isoiovia petraea growing in clefts and between gneissic boulders. 
Broken Hill. 

Fig. 6. — Prostanthera striaiiflora with Aristida and Ptilotus obovaiiis in right 
foreground, illustrating the nature of the lower shrubs and undergrowth of the 
Mulga scrub association. Broken Hill. 

Plate xvii. 

Fig. 7. — Low limestone hill with scattered mallee (Eucalyptus Gillei) ; ground 
flora made up entirely of species of Kochia and Bassia with Zygophyllum fruticu. 
losuni at the base of the trees. Corona. 

Fig. 8. — Slate outcrops near Old Corona on main stock route. Sparse vege- 
tation of Kochia sedifclia and K, pyramidata with ground covering of the grass, 
Schismus fasciculatus . Rubble sheet of fragmental quartz uncolonised. 

Fig. 9. — Limestone slope near Corona v/ith open community of species of 
Kochia. The trees near summit of slope are Casuarina lepidophloia. 

Plate xviii. 

Fig. 10. — Mi:lga-dotted slopes near Corona with old mulga in foreground in- 
fected with Uromycladium. 

Fig. 11. — Solitary tree oi Flindersia maculosa (Leopard- wood) with ground flora 
of Kochia sedifolia. Between Purnamoota and The Paps, Barrier Range. 

Fig. 12. — Flooded clayey flat behind creek showing colonisation by species of 
Bassia. 

Plate xix. 

Fig. 13. — Level sandy ground showing plants of Clianthus Davipiei i ('Sturt's 
Desert Pea) Broken Hill. 

Fig. 14. — Single plant of Clianthus Dampieri indicating prostrate habit and 
manner of growth. 

Fig. 15. — Bassia paradoxa colonising bare clayey soil between flood levels on a 
sandy creek. 

Plate XX. 

Fig. 16. — Depression in the Range between Broken Hill and Thackaringa — the 
Pinnacles in the distance — colonised chiefly b}^ Atriplex vesicarium with Kochia 
sedifolia on higher ground. 

Fig. 17. — Atriplex vesicarinm (ISaltbush) plain with a few trees of Acacia sp. 
and gum^creek (Eucalyptus rostrata) winding in distance. 

Fig. 18. — Typical barren hill close to Broken Hill. Kochia pyramidata, Kochia 
triptera and K. sedifolia and a few species of Bassia represent the only vegetation 
on the slopes and level ground and indicate dearth of annual vegetation during the 
1922 season. Note uncolonised rubble area. 



266 VEGETATION OF ARID AND SEMI-ARID NEW SOUTH WALES i. 

Fig. 19. — Saltbush plain with Kochia aphylla in the foreground and Rhagodia 
spinescens further away. Level ground between shrubs covered with close mat- 
like colony oi Marsilia Drummondii (Nardoo) . 

Plate xxi. 

Fig. 20. — Colony of Triodia irritans ("Porcupine grass" or "spinifex") near 
Stephen's Creek, Broken Hill. 

Fig. 21. — Single plant of Triodia irritans showing the formation of a typical 
hummock by the retaining of sand. 

Plate xxii. 

Fig. 22. — Stephen's Creek, near Broken Hill showing sandy bed and tall gums 
Eucalyptus rostrata). 

Fig. 23. — Clay-pan lake bordered by black Box fiat (Eucalyptus bicolor). 
Langawirra. 

Plate xxiil. 

Fig. 24. — Cobham Lake, a large clay-pan north of Broken Hill. Drought- 
stricken trees of Eucalyptus bicolor in foreground. 

Fig. 25. — Dry floor of Bancannia Lake, a clay-pan 100 miles north of Broken 
Hill — showing colonisation by Bassia divaricata and Chenopodium nitrariaceuni. 



267 



A MONOGRAPH OF THE FRESHWATER ENTOMOSTRACA OF 
NEW SOUTH WALES. Part iii. Ostracoda. 

By Marguerite Henry^ B.Sc, Linnean Macleay Fellow of the Society in Zoology. 

(Plates xxiv.-xxix.) 
[Read 27th June, 1923.] 



Introduction. 



The first description of Ostraeods collected in Nev/ South Wales was pub- 
lished in 1855 by the Rev. R. L. King who described briefly thirteen new species 
and a new genus, Newnhamia. In 1886 G. S. Brady described eight species, four 
of which were new, from the Tweed River district; these descriptions, as well as 
Kings, were based entirely on external features. In 1896 Sars described four 
new species and redescribed some of the earlier species. One new species was 
described by the present writer in 1919. The present paper comprises thirty-six 
species; two of these are recorded for the first time in Australia, one for the 
first time in New South Wales and seven are described as new. 

The records of Ostraeods present in the other States are very scanty. In 
Queensland, seven species were s^ecorded by Sars in 1889; with one exception 
these occur also in New South Wales. 

Brady described three species from South Australia in 1886 and two others 
are here recorded for the first time. Two species were described from Western 
Australia in 1896 by Sars. 

No records can be found of the species occurring in Victoria or Tasmania. 

The writer wishes to express her best thanks to Mr. T. Steel for the use of a 
collection of Ostraeods from England and Scotland for comparison with the Aus- 
tralian forms and for shells of two species of Candona from the Tweed River. 
Thanks are also due to Miss M. Collins, B.Sc, for samples of dried mud from 
the Cobar district, one of which yielded a new species of Cypridopsis, and also to 
many friends who have collected material. 

The drawings for this paper were all done with the aid of a camera lucida; 
the finished drawings were prepared by Miss D. Harrison. 

The type specimens of the new species have been deposited in the Australian 
Museum, Sydney. 

The folloAving lists give the Ostracoda recorded from the different States. 



268 FEESHWATER ENTOMOSTRACA OF N.S.W.^ iii., OSTRACODA^ 

Nexo South Wales. 
Cyprididae. 

Subfamily Notodeomadinae. — Notodromus fuscatus Brady, Newnhamia fene- 
strata King. 

Subfamily Ilyocyprinae. — Ilyocypris australiensis Sars. 

Subfamily Cypridopsinae. — Cy2Jridopsis ftmehris Brady, Cypridopsis aus- 
tralis, n.sp. 

Subfamily Cypridinae. — Cypris hennelong Eling, C. reticulata Zaddach, C. 
clarkii King, C. scottii King, C. lateraria King, Cypris crinita, n.sp., CyprinotuSt 
dentato-marginatus (Baird), Gy. fuscm Henry, Cy. incongruens Ramdohr, Cy. 
carinatus (King), Cy. leanus Sars, Cy. tenuis, n.sp., Stenocypris maleolmsonii 
(Brady), Cypretta minna (Kling), Cypretta globulus Sars, Cypretta turgida Sars, 
Cypretta viridis Thomson, Cypretta hirsuta, n.sp. 

Subfamily Cyclocypridinab. — Cypria pusilla Sars, Cyclocypris tenuissima, 
n.sp. 

Subfamily Herpeoxdcypridinae. — Candonocypris candonoides (King), Her- 
petocypris laevissima, n.sp., IlyodrOmus varrovillius (King), I. viridulus Brady^ 
I. suhstriatus Sars, I. elliptieus Saxs, I. ohtvsus Sars, I. stOMleyanus (King). 

Subfamily Cando^s^HnTAB. — Candona lutea King, Candonopsis tenuis (Brady). 
Cytheridae. — Limnicythere aspera, n.sp. 

Queensland. 
Cyprididae. 

Subfamily Cypridinae. — Cyprinotus dentato-marginatus (Baird), Cyprinotus 
cingalensis Brady, Cypretta globulus Sars, Stenocypris maleolmsonii (BradjO- 

Subfamily Herpetocypridinae. — Candonocypris candonoides (King), Ilyo- 
dromn-s viridulus Brady. 

Subfamily Ilyocyprinae. — Ilyocypris australiensis Sars. 

South Australia. 
Cyprididae. 

Subfamily Cypridinae. — Cypris tatei Brady, C. mytiloides Brady, C. benne- 
long KJing, Cypretta viridis Thomson. 

Subfamily Herpetocypridinae. — Ilyodromus stanleyanus (King). 

Western Australia. 
Cyprididae. 

Subfamily Cypridixae. — Amphicypris oblongata Sars, Cyprinotus dahli Sars, 

Classification. 

The Ostraeoda are divided into four great tribes; three of these possess a 
biramose antenna variously developed and, so far as is known, are exclusively 
marine. The tribe Podocopa comprises all the freshwater forms and, in addition, 
a large number of marine forms; they are distinguished by the possession of n 
simple antenna which always bears apical claws. 

Key to families of Podocopa. 
A. Two pairs of legs; antenna without a flagellum. 

B . AlDdomen without a furca Dnrrcinulidae. 

(not known in Australia). 



BY MARGUERITE HENRY. 269 

BB , Abdomen with a furca. . . . . Cyprididae. 

AA. Three pairs of legs; antenna with a flagellum Cytheridae. 

Family CYPRIDIDAE. 

Shell usually thin, surface smooth, striated or pitted. Antennules com- 
posed of seven segments. Antennae composed of four to six segments, without 
a flagellum, with or without a brush of swimming setae. Two dissimilar pairs of 
legs. Caudal rami usually well developed, sometimes rudimentary. Seven sub- 
iamilies are represented in New South Wales. 



OS 



Key to subfamilies of Cyprididae. /^ ^ oi^, v-. 

/ ^ 
A. Antennae with natatory setae. I^jj L i B ?? •■' 

B. Natatory setae long, at least reaching the tips of the terminal claws. \-^ \ -^^f* 
C. Natatory setae reaching the tips of the claws or slightly beyond. Y'^/K 

D. Second foot with three terminal setae of different lengths. \P'^/"^<?A: 

E. Two eyes present Notodromadinae. x^^ ^ 

EE. One eye Ilyocyprinae. 

DD. Second foot with a beak-like end segment and a claw. 

E. Furca rudimentary. .. Cypridopsinae. 

EE. Furca normal Cypridinae. 

CC. Natatory setae exceeding the terminal claws by half their length. .. 

Cycle cypridinae. 
BB. Natatory setae shortened, not nearly reaching the tips of the terminal 

claws Herpetocypridinae. 

AA. Antennae without natatory setae Candoninae. 

Subfamily NOTODROMADINAE 

Shells usually short and high. Natatory setae of the antennae slightly ex- 
ceeding the tips of the terminal claws. Second leg with three terminal setae of 
different lengths; one seta reflexed. Two eyes present. 

Two genera are represented in New South Wales. 

Key to genera of Notodromadinae. 

A. Second maxilla with two branchial filaments attached directly to the limb. 

Newnhamia. 
AA. Second maxilla without branchial filaments Notodromus. 

Genus Notodromus Lilljeborg, 1853. 

Shell short and high, with a ventral flattening. Antennae composed of six 
segments. Furca with three seta-like claws, terminal seta absent. Both sexes 
present. Six species are known; one occurs in New South Wales. 

Notodromus fuscatus Brady. 

This species was described by Brady in 1886 (p. 92, Plate x., figs. 4-6) but 
since merely the external appearance was described it is impossible to decide 
whether it is a true Notodromus or a member of the allied getous Newnhamia. 
Brady's description is as follows. "Shell seen laterally sub-triangular, height 
equal to |:liree-fourths the length, extremities broadly rounded, anterior narrower 
of the two, dorsal margin excessively arched, highest a little behind the middle; 



270 FRESHWATER ENTOMOSTRACA OF N.S.W., iii., OSTRACODA, 

ventral nearly straight; seen from above the outline is ovate, scarcely twice as 
long as broad, tapered and acuminate in front, rounded off behind. Surface of 
the valves somewhat rough and furfuraceous, colour brownish with darker cloud- 
ings." 

Distributiofi. — Brady's specimens were collected in the Condong District, 
Tweed River, N.S.W. 

The late Mr. G. I. Playfair made drawings of specimens from Lismor©, 
N.S.W. , which seem to be identical with those from the Tweed. 

Genus Newnhamia King, 1855. 

Shell granulate or tuberculate, appendages like those of Notodromus except 
that the mandibular palp bears a small branchial appendage, the filaments of 
which are directed upwards; the second pair of maxillae have two branchial fila- 
ments attached directly to the limb. Propagation sexual. Only two species are 
known, N. patagonica Vavra from Patagonia and N. fenestrata King from Aus- 
tralia. 

Neivnhamia fenestrata King. (Plate xxiv., figs. 1-10.) 

King, Proc. Roy. Soc. Van Diemen's Land, 1855, p. 67. 

Female. Shell seen laterally (fig. 1) short and high, the greatest height oc- 
curring about the middle; dorsal margin ajched rather abruptly behind the eye, 
ventral straight for the great part of its length, becoming rounded off at each 
end; anterior and posterior margins slightly curved. Viewed dorsally (fig. 2), 
the outline is broadly ovate, tapering anteriorly with the greatest width, which 
exceeds two-thirds of the length, situated behind the middle. Ventral surface 
with a long flattened ventral plate (fig. 5). Valves unequal, the right slightly 
larger than the left. Surface thickly covered with tubercles, the margins bearing 
scattered hairs (figs. 3-4). Caudal rami (fig. 7) bearing three long seta-like 
claws, the two terminal ones almost equal in length, terminal seta absent. Colour 
dark grey to brown. Length, .79 mm., height, .61 mm. 

Male. Smaller than the female, shell differently shaped, the highest point 
being behind the middle. 

Distribution. — This species is common in the neighbourhood of Sydney; it is 
usually found in permanent ponds and in great numbers. N.S.W. : Sydney, 
Narrabeen, Lane Cove, Liverpool, Moss Vale, Holbrook. 

Brady recorded it from New Zealand in 1906 and incorrectly stated that 
icing's specimens were from Tasmania. Vavra also found it in the Bismarck 
Archipelago. 

Subfamily ILYOCYPRINAE. 

Shell strong, surface rough, with grooves and tubercles resembling the mem- 
bers of the family Gytlieridae. Setae of antennae of varying lengths, usually ex- 
tending at least to the tips of the terminal claws. Second leg with three setae 
on the teiminal segment, one seta reflexed. One genus, Ilyocypris. 

Genus Ilyocypris Brady and Norman, 1889. 

Shell compressed, surface pitted or tuberculated. Second pair of maxillae 
with a Avell developed branchial lamella. First ieg only composed of five seg- 
ments. Caudal rami with the dorsal seta attached almost in the middle. Pro- 
pagation sexual. Seventeen species have been neRcril)ed, one of which occurs in 
New South Wales. 



BT MARGUERITE HENRY. 271 

Ilyocypris australiensis Sars. 

Described in detail by Sars (1889) and fully flgnired (Plate ii., figs. 5-8; PI. 
vi., figs. 1-14). 

Distribution. — Sars' specimens were raised from dried mud collected at the 
Graeemere lagoon, near Rockhampton, Queensland. A few specimens were found 
in a collection from Holbrook. It has not hitherto been recorded from New 
South Wales. 

Subfamily CYPRIDOPSINAE. 

Natatory setae of the antennae reaching to the tips of the terminal claws. 
Second leg beak-shaped at the tip, with a terminal claw. Furea rudimentary, 
terminating in a long seta. 

One genus represented in New South Wales. 

Genus Cypridopsis Brady, 1868. 

Shell tumid. Antennae composed of five seg-ments. Furca rudimentary, 
ending in a long seta and with a short seta on the dorsal edge. 

Several species have been described from Australia and New Zealand as be-- 
longing to this genus, but in all except one case they were found to possess a 
normally developed furca. 

Cypridopsis funehris Brady is a doubtful species. The first certain species 
of Cypridopsis in Australia is here recorded for the first time. 

Cypridopsis australis^ n.sp. (Plate xxv., figs. 1-7.) 

Female. Seen laterally (fig. 1) somewhat oval in outline; dorsal margin 
boldly arched, with a longer slope anteriorly than posteriorly; ventral margin dis- 
tinctly sinuated just behind the middle; anterior and posterior margins evenly 
curved, the anterior lower than the posterior; gTeatest height occurring in front 
of the middle. Seen dorsally (fig. 2), elongated oval, slightly more pointed an- 
teriorly than posteriorly; greatest width equal to half the length and occurring 
about the middle. Valves thick and strong, slightly unequal, the left being larger 
than the right. Surface irregmlarly marked by rounded pits (fig. 3) and bearing 
scattered hairs. Antennules with moderately long swimming setae. Antennae 
(fig. 4) with the setae of the second last segment exceeding the length of the 
strong terminal claws. First leg (fig. 5) with a strong terminal claw the distal 
half of which is denticulated. Second leg (fig. 6) bearing a straight tenninal 
claw and a slender seta, Furca (fig. 7) rudimentary, with a small dorsal seta 
and a long terminal flagellum. Colour brown, becoming grey in alcohol. Length, 
.61 mm., height, .37 mm. 

Distribution. — This species was bred from dried mud collected at Meryula 
station, near Cobar, N.S.W. 

Cypridopsis punebris Brady. 

This species was described by Brady in 1886 (p. 91, Plate viii., figs. 7-9). 
The description is only of the shape of the shell which appears to have been im- 
mature. Without examining the furea it is impossible to state whether this form 
is a true Cypridopsis and it is even more doubtful since Brady includes Cypretta 
minna King in this genus. The specimens were collected in the Condong dis- 
trict, TAveed River, N.S.W. 



272 FRESHWATER ENTOMOSTRACA OF N.S.W.^ iii., OSTRACODA, 

Subfamily CYPRIDINAE. 
Natatory setae of the antennae reaching at least to the tips of the terminal 
claws. Second leg with a beak-like end segment and claw. Furca well developed, 
usually armed with both claws and setae. Three genera are represented in New 
South Wales. 

Key to genera of Cypridinae. 

A. Furca with two long terminal setae replacing the usual claws Cypretta. 

AA. Furca with claws and setae. 

B. Dorsal seta of furca rudimentary or absent Stenocypris. 

BB. Dorsal seta well developed. Cypris. 

Genus Cypretta Vavra, 1895. 

Shell short and tumid. Natatory setae of the antennae reaching beyend 
the terminal claws. Furca with the terminal claws replaced by setae, usual ter- 
minal seta sometimes absent. Ovary spirally wound. Males unknown. Five 
species occur in New South Wales. 

Key to species of Cypretta. 

A. Surface of the shell distinctly marked by closely set pits. 

B. Furca with three long terminal setae glohulus. 

BB. Furca with two long setae hirsuta. 

AA. Surface of the shell smooth or with a few scattered pits. 

B . Right valve projecting over the left dorsally minna. 

BB. Valves even dorsally. 

C. Width not exceeding three-quarters of the length viridis. 

CC. Width almost equal to the length turgida. 

Cypretta globulus Sars. 

Cypridopsis glohulus Sars, 1889 (PI. ii., figs. 9, 10; PL vii., figs. 1-11). 

Distribution. — N.S.W. : Five Dock, Kensington, Kendall, Holbrook; Queens- 
land. 

The average length of this species is .75 mm. Some very much larger 
specimens were obtained from Five Dock; these attained a length of .90 mm., 
but were otherwise identical with the smaller forms. 

Cypretta hirsuta^ n.sp. (Plate xxviii., figs. 1-5.) 

Female. Seen laterally (fig. 1) irregularly oval in outline, the greatest 
height exceeding two-thirds of the length and occurring just behind the middle; 
dorsal margin evenly arched, ventral straight, anterior rounded, posterior lower 
than the anterior and almost angular. Seen dorsally (fig. 2) broadly oval in 
outline, greatest width slightly greater than the height and occurring about the 
middle; anterior end obtusely pointed, posterior rounded. Surface of the shell 
distinctly pitted and everywhere densely hairy. Right valve slightly overlapping 
the left, the ends marked with transverse lines (fig. 3) ; Muller (1898) regards 
these as septa which are extended between the valve lamellae for support; they 
appear to be characteristic of the genus and are especially distinct in this species. 
Mouth parts typical of the subfamily. ^Second pair of legs (fig. 4) provided 
with a hook and a slender recurved seta. Caudal rami long and slender, apical 



BY MARGUERITE HENRY. 273 

seta usually long', second seta not attaining half the length of the first; both 
apical and dorsal minute setae present. Colour dark green with two lighter 
patches. 

Leng-th, .88 mm., height, .61 mm., width, .63 mm. 

Distribution.— l>i .S.W ., Kosciusko. 

Cypretta turgida Sars. 

Described by Sars in 1894 (Plate iv., fig. 3a-d) as Cypridopsis minna and 
renamed Cypridopsis turgida in 1896. 

Female. Shell, seen laterally, somewhat semicircular in shape, the greatest 
height occurring behind the middle and equal to two-thirds the length; dorsal 
margin boldly arched, sloping more steeply posteriorly than anteriorly; ventral 
margin straight, anterior and posterior ends rounded. Viewed dorsally, the an- 
terior end is narrower than the posterior but the general shape is almost circular, 
the greatest width nearly equalling the length. Valves unequal, the right being 
slightly larger than the left and overlapping it anteriorly. Surface smooth, 
with a few small pits, hairs almost confined to the extremities. Caudal rami long 
and narrow, with terminal claw-like setae. Colour light yellowish-brown. Length, 
.90 mm. 

Distribution. — N.S.W. : Sydney, Moss Vale, Berrima, Holbrook; New Zea- 
land; China; Madagascar; Sumatra. 

Cypretta viridis Thomson. 

Described in 1878 (Plate vi., figs. A2a-9) as Gypris viridis. 

Female. Seen laterally, rounded oval in form; dorsal margin arched, slop- 
ing steeply anteriorly; ventral margin straight; anterior margin lower than the 
posterior. Seen dorsally, oval in outline with the anterior end narrower than 
the posterior; width slightly greater than the height. 

Valves unequal, the right being larger than the left. Surface smooth with 
a few scattered pits; densely covered v/ith hairs. Caudal rami slender, with long 
setae. Colour dark green vifith irregular patches of a still darker shade. Length, 
.95 mm. 

Distribution. — N.S.W. : Botany; South Australia; New Zealand. 

Cypretta mink a (King). 

Described as Cypris minna in 1855 (p. 64) ; transferred to the genus Cypri- 
dopsis by Brady in 1886; described and figured by Sars in 1896 (Plate vii., fig. 
5a-e) as Cypridopsis minna. 

Female. Shell, seen laterally, rounded triangular in shape, the greatest 
height occurring about the middle and almost equal to the length; dorsal margin 
angular in the middle, sloping steeply to either side, ventral margin sinuated in 
the middle. Seen dorsally, somewhat egg-shaped, the greatest width not quite 
equalling the height. Valves unequal, the right overlapping the left anteriorly 
and dorsally. Surface slightly granular and hairy. Caudal rami very nan'ow, 
setae long. Length: Sars gives .92 mm. as the greatest length attained, but 
many specimens have been found measuring as much as 1.0 mm. and this is 
accordingly the largest Australian Cypretta. Colour, yellowish with large vivid 
gxeen patches which become peacock blue in alcohol. 

Distribution. — N.S.W. : Sydney, Varroville, Denham Court, Centennial Park, 



274 FRESHWATER ENTOMOSTRACA OF N.S.W.^ iii,, OSTRACODA^ 

Lane Cove, Condong district, Tweed River. Many specimens were bred out of 
dried mud collected at Meryula station, near Cobar. 

Genus Stenocypris Sars, 1889. 

Shell narrow, the height not nearly attaining half the length. Natatory 
setae of the antennae not exceeding the tips of the terminal claws. First maxilla 
with a narrow cylindrical palp, last joint very small; masticatory lobes long and 
narrow. Caudal rami large, claws coarsely denticulate, dorsal seta absent or 
very small. Propagation sexual. 

Nineteen species have been described, one of which occurs in New South 
Wales. 

Stenocypris malcolmsonii (Brady). 

Syn. Cypris cylindrica Baird. 

Described in 1886 (p. 297) as Cypris malcolmsonii and made the tj^pe of a 
new genus by Sars in 1889 (PI. i., figs. 7, 8; PL v., figs. 1-4) . 

Female. Shell, seen laterally, elongated and narrow, height uniform, not 
nearly attaining half the length; dorsal margin straight for the greater part of 
its length, sloping abruptly in front and behind; ventral margin distinctly 
sinuated in front of the middle; anterior and posterior margins rounded. Seen 
dorsally, very narrow, with straight sides, anterior end more pointed than pos- 
terior. Valves almost equal, the left slightly longer than the right, inner dupli- 
eatures broad. Caudal rami strong, unequal, the right broader than the left 
and bearing a row of strong denticles, left ramus tapering slightly and bearing 
few small denticles; both claws coarsely denticulate, dorsal seta absent. Length, 
1.7 mm. Colour pale green. 

Distribution. — N.S.W. : Casino; Queensland; India; Ceylon. 

Genus Cypris Muller. 

Shell of various shapes. Antennae composed of five segments, natatory 
setae reaching to the tips of the terminal claws or beyond. Furca well developed, 
bearing claws and setae. Propagation sexual or parthenogenetic. 

This large genus has been divided into a number of subgenera; two are 
represented in New South Wales. 

Key to subgenera of Cypris. 

A. Inner edge of the right shell thickly tuberculate Cyprinotus. 

AA. Inner edge plain Cypris. 

Subgenus Cypris 

First leg five-segmented. Inner edge of the right valve not tuberculate. 
Furca normal with two claws and two setae. 

Key to species of subgenus Cypris. 

A. Length not attaining 3 mm. 

B. Shell with a prominent anterior flange bennelong. 

BB. No such flange. 

C. 'Surface of the shell sculptured. 

D. Shell armed with tubercles lateraria. 



BY MARGUERITE HEJNRY. 275 

DD. Shell without tubercles reticulata. 

CC. Surface of the shell smooth .. crinita. 

AA. Length exceeding 3 mm scottii. 

Cypris bennelong King. 

Syn. Chlamydotheca amtralis Brady. 

Described very briefly by King in 1855 (p. 63) from an immature specimen. 
In 1894 (PI. iv., figs, la-d) Sars described the same species from New Zealand 
and, on examining immature forms, decided that it was identical with the form 
described by King. Although the shell has the flange that is characteristic of 
the genus Chlamydotheca, the structure of the oral parts shows it to be a true 
Cypris, the natatory setae of the antennae extending well beyond the tips of the 
terminal claws. 

Female. Shell, seen laterally, oval triangular, the greatest height exceeding 
half the length and occurring about the middle; dorsal margin arched, sloping- 
more steeply anteriorly than posteriorly; ventral margin with two slight sinuses; 
anterior margin rounded, projecting into a flange at the lower corner, the pro- 
minence being caused by the left A^alve considerably overlapping the right. Seen 
dorsally the outline is oblong, with the greatest width equal to half the length, 
sides nearly straight; anterior end narrower than posterior, with a peculiar twist 
formed by the projection of the left valve. Surface smooth, margins of the 
valves hairy. Natatory setae of the antennae projecting far beyond the terminal 
claws. Caudal rami long and slender, apical claw equalling half the length of 
the ramus. Length, 1.4 mm. Colour brownish-green. 

Distribution. — N.S.W. : Sydney Cove, Bourke Street, Corona; South Aus- 
tralia; New Zealand. 

Cypris lateraria King. 

Described by King in 1855 (p. 65, Plate x., G.) 

Female. Shell, seen laterally, much higher anteriorly than posteriorly, the 
greatest height slightly greater than half the length; anterior margin rounded; 
posterior obliquely truncated; dorsal arched above the eye, thence sloping in a 
straight line to the posterior margin; ventral margin deeply sinuated about the 
middle. Seen dorsally the shell appears oval, more pointed anteriorly than pos- 
teriorly, the greatest breadth equalling the height. Left valve larger than the 
right, overlapping it anteriorly and posteriorly; inner duplicatures narrow. Sur- 
face of the shell granular and densely hairy, bearing tubercles. Caudal rami 
slightly narrowed towards the tips, the outer claw not attaining half the length 
of the ramus. Colour yellowish-green. Length, 1.05 mm., height, .63 mm. 

Distribution. — N.S.W. : Bourke Street, Sydney, Bringagee. 

This species was reared from dried mud from Bringagee and from Corona. 

Cypris reticulata Zaddach. 

Zaddach, Synopseos crustaeeorum Prussieorum Prodromus, 1844. p. 34. 

Female. Shell, seen laterally, with the gTeatest height occurring in front of 
the middle; dorsal margin arched especially in the anterior portion; ventral 
margin with a slight sinuation behind the middle, anterior and posterior margins 
rounded. Seen dorsally, broadly oval, the greatest width occurring about the 
middle, anterior end narrower than the posterior. Natatory setae of the an- 
tennae reaching slightly beyond the tips of the terminal claws. Surface of the 
shell sculptured with a fine reticulation which is more distinct in younger speei- 



276 FRESHWATER ENTOMOSTRACA OF N.S.W.^ iii., OSTRACODA^ 

mens. Caudal rami bent near the apex, terminal claws denticulate, terminal and 
dorsal setae of equal length. Length, 1.3 mm. Colour yellowish to brown with 
dark markings. 

Distribution. — N.S.W. : Holbrook; North America; Europe. 

This is the first record of the occurrence of this species in Australia. Tlie 
specimens examined agreed exactly with the descriptions of European forms, the 
only difference being a darker coloration. 

Cyfris crinita, n.sp. (Plate xxvi., figs. 1-8.) 

Female (fig. 1). Shell, seen laterally, irregularly oval in outline, the 
greatest height exceeding half the length and occurring slightly in front of the 
middle; dorsal margin arched, sloping evenly at each end; ventral margin gently 
curved, sinuated in the middle; anterior and posterior margins evenly curved, 
the anterior lower than the posterior. Seen dorsally (fig. 3) a regular oval, 
except at each end, where there is a projection of the left valve; greatest width 
occurring about the middle and not quite equalling the height. Valves unequal, 
the left overlapping the right anteriorly, posteriorly and ventrally. Surface 
smooth, except for the usual small scattered pits, anterior and ventral margins 
bearing hairs, those of the posterior margin being unusually long. Antennules (fig. 
4) with moderately long setae. Antennae (fig. 5) with the setae of the ante- 
penultimate segment easily exceeding the tips of the terminal claws. First leg 
(fig. 6) with a powerful curved terminal claw bearing a row of denticles. Caudal 
rami (fig. 8) very long and exceptionally slender, terminal claw almost as long 
as the ramus, second claw only one-third as long as the terminal one, both claws 
denticulate and straight except for a pronounced hook at the end; terminal seta 
slightly shorter than the dorsal. Colour brilliant green and orange. Length, 
2.2 mm., height, 1.2 mm., width, 1.1 mm. 

Blale (Fig. 2). Smaller than the female, the largest found only attainiiig 
a length of 1.9 mm. Shape somewhat different from that of the female, the 
height being greater in comparison with the length, the dorsal margin more 
pronouncedly arched and the ventral margin straighter. 

Distribution. — N-S-W. : Holbrook. 

Cyfris soottii King. 

This species has not been recorded since King very briefiy described it in 
1855 (p. 63, Plate x., fig. C). It is evidently distinct from any other species 
recorded in Australia, since King describes it as being "nearly the tenth of an 
inch in length, of a transparent green colour marked with very minute reddish 
spots." The locality given is Denham Court. 

Cyfris stobarti King. 

Described by King (1855, p. 62, Plate ix.. Pig. B) from a single specimen 
which was not dissected. It is impossible to decide from the description whether 
this is a true Cypris; its outline suggests the subfamily Herpetocypridinae. 
King's description is as follows: — "The shell is oblong and slightly sinuated on 
the posterior dorsal margin. The valves are unequal, the left being the larger. 
They are polished and apparently of a yelloAvish colour." 

Distribution. — Queensland : Moreton Bay. 



BY MARGUERITE "HENRY. 277 

Cypris CLARKII KiBg. 

This species also suggests the Herpetocypridinae but until further specimens 
are obtained it is impossible to decide whether it is a true Cypris. King's 
description (1855, p. 63, Plate x., E) only gives colour and form. "Shell oval, 
slightly reniform, the valves very convex, variegated with brown and a light 
reddish green in well defined notches of irregular but constant shape; the eye 
is yellow, shell punctured pilose." 

Distribution. — N.S.W. : Sydney, Parramatta. 

Subgenus Cyprinotus Brady, 1885. 

Shells usually high, the greatest height being more than half the length; 
inner margin of the right valve thickly tuberculate. Brady described Cyprinotus 
as a new genus in 1885 (p. 301) with C. cingalensis as the type species, the in- 
equality of the valves being taken as a generic character; in this species the 
right valve is gibbous and overlaps the left valve dorsally. In 1889 Sars 
described C. dentato-marginatus from Queensland; this species obviously belongs 
to the same genus as C. cingalensis, but lacks the dorsal projection of the right 
valve. Sars therefore based the genus on the tubereulate right valve and the 
fact that the propagation was sexual as contrasted with the genus Cypris which, 
was supposed to be exclusively parthenogenetic. In 1903, Sharpe reduced 
Cyprinotus to a subgenus of Cypris, distinguished by the possession of tubercles 
on the right valve margin. He pointed out that the method of propagation is 
not always a generic character among the Ostraeoda and that there are genuine 
species of Cypris, such as C. testudinaria, which propagate sexually. In this 
paper two species described by Sars under the genus Cypris have been trans- 
ferred to the subgenus Cyprinotus ; their males are unknown but there is no proof 
that they are exclusively parthenogenetic althovigh they are known to propagate 
in this manner under certain circumstances. 

Key to species of the subgenus Cyprinotus, ' 

A. Surface smooth and polished. 

B. Right valve forming a dorsal projection. 

C. Right valve considerably overlapping the left ventrally fuscus. 

CC. Valves even ventrally carinata. 

BE. No dorsal projection of the right valve. 

C. Length exceeding 2 mm. .. .. leana. 

GC. Length not attaining 2 mm. 

D. Greatest height occurring in the middle dentato-marginatus. 

DD. Greatest height occurring posterior to the middle. incongruens. 
AA. Surface marked by a reticulate pattern tenuis. 

Cyprinotus fuscus Henry. 

Proc. Roy. Soe. N.S.W., liii., 1919, p. 44, Plate ii., figs. 13, 14. 
Distribution. — N.S.W. : Botany, Lismore. 

Cyprinotus carinata King. 

Desci-ibed in 1855 (p. 61, PI. ix., figs. Cl-4) as Cypris carinata. The brief 
description and the figures are characteristic of this subgenus; the shape of the 
shell somewhat resiembles C. cingalensis, although the dorsal projection is situated 



278 I'HESHWATKR ENTOMOSTRACA OF N.S.W.^ iii., OSTRACODA^ 

farther back on the shell. King's description is as follows: — "Shell nearly 
elliptical, but higher on the back; ^he valves are unequal, the right being pro- 
duced beyond the left at the posterior part of the dorsal edge giving the shell 
the appearance of a bell; the valves are polished, of a transparent greenish 
colour, with a darker qiTadrangular mark in the middle. Males darker and 
somewhat smaller." 

Distribution. — Denham Court. 

Cyprinotus leana Sars. 

Described in 1896 (Plate vii., figs. 2, a-c) as Cypris leana. 

Female. Shell, seen laterally, oval reniform, the . greatest height occurring 
at the middle; dorsal margin greatly arcbed; ventral almost straight; anterior 
margin rounded, posterior higher than the anterior and obliquely truncated. 
Seen dorsally, oval in shape, the greatest width occurring behind the middle and 
equalling half the length; more pointed anteriorly than posteriorly. Valves 
slightly unequal, the right valve bearing a row of tubercles on the anterior margin 
and on the posterior portion of the ventral margin. Inner duplieatures narrow. 
Surface of the shell smooth and polished, the ends bearing small hairs. Candal 
rami slendea", tapering distally, apical claw equalling half the length of the 
ramus. Colour yellowish-brown. Lengih, 2.7 mm. 

Distrihution. — Hay, Yass. 

Cyprinotus dentato-marginatus (Baird). 

Described in 1859 (p. 233) as Cypris dentato-marginatus. Fully described 
and figured in detail by Sars in 1889 (PI. i., figs. 1-4, PI. iii., figs. 1-11; PI. iv., 
figs. 1-14). 

Distribution. — N.S.W. : Botany, ,Centfennial Park; Queensland; India; 
Ceylon. 

Cyprinotus incongruens Ramdohr. 

Syn. C. sydneia Bang, C. ciliata Thomson. 

Fii'st described by Ramdohr in 1808 (p. 83) ; described by King in 1855 (p. 
65) as Cypris sydneia and figured by Sars under that name in 1894 (Plate iv., 
figs. 2 a-c). 

Female. Shell, seen laterally, reniform, higher posteriorly than anteriorly, 
the greatest height c--curring behind the iniddle and exceeding half the length; 
dorsal margin curved, with a longer slope in front than behind; ventral margin 
almost straight. Seen dorsally, somewhat oval, much broader posteriorly than 
anteriorly, breadth not as great as the height. Left valve larger than the right, 
overlapping it at each end and ventrally; right valve tuberculated along the an- 
terior margin and postero-ventraUy. Surface smooth and polished with a few 
scattered pits. Caudal rami curved, apical claw not nearly half as long as the 
ramus. Colour yellow to orange. Length, 1.4 mm. 

Distribution. — This is one of the few Ostracods with a world-wide distri- 
bution. It is common in the neighbourhood of Sydney and specimens have been 
collected at Bangalow on the North Coast of New South Wales. It has been re- 
corded from New Zealand, North America, Europe and Asia. 

Cyprinotus tenuis, n.sp. (Plate xxvii., figs. 1-8.) 
Female (fig. 1). Seen laterally, irregularly oval in outline, with the 



BY MARGUERITE HENRY. 279 

greatest height occurring in front of the middle. Dorsal margin arched, sloping 
more abruptly anteriorly than posteriorly; ventral margin straight; anterior and 
posterior margins truncated, the anterior higher than the posterior. Seen dor- 
sally (fig. 2) elongated oval, the greatest width occurring behind the middle, the 
sides curving to the posterior end but tapering to the more pointed anterior end. 
Valves unequal, the left being larger than the right; right valve bearing a row 
of denticles on its anterior margin and also on the postero-ventral corner; inner 
duplicatures narrow. Surface of the shell covered with an irregular reticulate 
pattern, with diamond-shaped meshes enclosing numerous small pits (fig. 3). 
Antennules (fig. 5) bearing long swimming setae, those of the antennae (fig. 4) 
reaching beyond the terminal claws. Second leg (fig. 7) bearing a short curved 
claw and a long seta. Caudal rami (fig. 8) comparatively short, terminal claw 
equal to three-quarters the length of the ramus, second claw slightly shorter, 
neither claw bearing denticles, apical seta short, only attaining half the length of 
the dorsal seta. Colour gi'eyish brown, very transparent in alcohol. Length, 1.1 
mm., height, .59 mm. 

This species somewhat resembles C. pellucida Sharpe; an important differ- 
ence is that in C. pellucida the right valve is larger than the left, whereas in the 
present form the exact opposite occurs. 

Distribution. — N.S.W. : Eensington. 

Subfamily CYCLOCYPRIDINAE. 

Natatorj^ setae of the antennae very long, exceeding the terminal claws by 
about half their lengih. Second leg with three setae of different length, one or 
more reflexed. Furca normal. 

Key to genera of Cycloeypridinae. 

A. Terminal segment of the second leg small, with two short setae and a long 

reflexed seta Cypria. 

AA. Terminal segment of the second leg long, with one short claw and two re- 
flexed setae ; Cyclocypris. 

Genus Cypria Zenker, 1854. 

Shell short and high, strongly compressed. Second antenna of the male 
with two sense organs on the fourth segment. Last segment of the second leg 
short, bearing two short claw-like setae and one long reflexed seta. One species 
occurs in New South Wales. 

Cypria pusilla Sars. 

Described in 1896 (Plate vii., figs, la, b). 

Female. Shell, seen laterally, somewhat semicii'cular in outline, the greatest 
height occurring about the middle; posterior margin rounded, higher than the 
anterior which is oblique; dorsal margin boldly arched; ventral almost straight. 
Seen dorsally, oblong, the anterior end slightly narrower than the posterior. 
Valves unequal, the right slightly overlapping the left anteriorly and postero- 
ventrally, and projecting much more dorsally. Surface of the shell smooth, 
margins densely hairy. Colour reddish-brown. Length, .58 mm. 

Distribution. — N.S.W. : Rarely found in stagnant pools in the vicinity of 
Sydney. Sars' specimens were collected at Waterloo. 



280 FEESHWATER ENTOMOSTRACA OF N.S.W.^ iii., OSTEACODA^ 

Genus Cyclocypris Brady and Norman, 1889. 

Second antenna of the male without sense organs on the fourth segment. 
Terminal segment of the second leg long and narrow, bearing a short claw-like 
seta and two long reflexed setae. About fourteen species have been described. 

Cyclocypris tenuissima^ n.sp. (Plate xxix., figs. 1-3.) 

Female (fig. 2). Shell, seen laterally, oval in outline, dorsal margin evenly 
curved; ventral distinctly sinuated behind the middle; anterior margin curved, 
higher than the posterior; greatest height occurring in front of the middle and 
equal to slightly more than half the length. Seen dorsally, broadly oval, more 
pointed anteriorly than posteriorly, greatest width occurring behind the middle. 
Valves slightly unequal, the left overlapping the right anteriorly and posteriorly. 
Surface granular, though not definitely sculptured, free margins sparsely hairy. 
Natatory setae of the antennules and antennae very long. Second leg with a 
moderately long end-segment bearing a short claw-like seta and two reflexed 
setae, one of which is nearly twice as long as the other. Caudal rami (fig. 3) 
long and slender, curved, end-claws long. Dorsal seta almost equal to the apical 
seta in length. Colour brown. Length, .52 mm., height, .28 mm. 

3Iale (fig. 1). Slightly smaller than the female and higher in proportion 
to its length; dorsal margin more boldly arched; ventral almost straight. Length, 
.50 mm., height, .30 mm. 

Bistrihution.- — N.S.W. : Orange. 

Subfamily HERPETOCYPRIDINAE. 

Natatory setae of the antennae shortened. Second kg terminating in a 
beak-shaped end-segment armed with a short claw. Incapable of swimming in 
the adult state. Three genera are represented in New South Wales. 

Key to genera of Herpetoeypridinae. 

A. Furca normal. 

B . Left valve larger than the right Kerpetocypris. 

BB. Right valve larger than the left Candonocypris. 

AA. Furca ending in three claws Ilyodromus. 

Genus Candonocypris Sars, 1894. 

Shell oblong, height not attaining half the length. Right valve overlapping 
the left. Natatory setae of the antennae not attaining the tips of the terminal 
claws. First pair of maxillae with short masticatory lobes, palp large, bearing 
a few claw-like spines. Propagation usiially parthenogenetic. 

One species occurs in New South Wales. 

Candonocypris candonoides (King). 

Described by King in 1855 (p. 66) as Cypris candonoides. Redescribed and 
figured in detail by Sars in 1889 (p. 35, Plate ii., figs. 1-2; PI. v., figs. 5-7) as 
Herpetocypris stanleyana. Transferred to a new genus Candonocypris by Sars 
in 1894. 

Distribution. — N.S.W. : Sydney, Varroville, Buckanbe; Queensland; New 
Zealand; South Africa. 



BY MARGUERITE HENRY. 281 

Genus Herpetocypris Brady and Norman, 1889. 

Shell elongated. Natatory setae of the antennae not attaining the tips of 
the terminal claws. Second segment of the first leg with one seta on the anterior 
margin. Maxillae as in Cypris. One species present in New South Wales. 

Herpetocypris laevissima^ n.sp. (Plate xxviii., figs. 6, 7.) 

Female (fig. 6). Seen laterally, oval in outline, with the greatest height, 
which slightly exceeds half the length, situated behind the middle. Dorsal margin 
arched, sloping more abruptly posteriorly than anteriorly; ventral margin dis- 
tinctly sinuated about the middle; anterior and posterior margins rounded, the 
anterior being lower than the posterior. Seen dorsally, regularly oval in outline, 
with the greatest width occurring about the middle, anterior end more pointed 
than the posterior. Valves equal, margins sparsely hairy. Surface of the shell 
smooth, except for a few scattered and very tiny pits. Antennules with moder- 
ately' long seta.e. Antennae with very short setae which do not reach beyond the 
base of the terminal claws. Second leg ending in a curved claw and bearing a 
long seta. Caudal ramus (fig. 7) long and slender, second claw attaining two- 
thirds the length of the apical claw, both minutely denticulated; setae of equal 
lengih; dorsal edge of the ramus bearing a row of minute denticles. Colour 
gTeenish-yellow. Length, 1.4 mrh., height, .72 mm. 

Male unknown. 

Distribution. — N.S.W. : Parramatta. 

Genus Ilyodromus Sars, 1898. 

Shells highly compressed. Valves equal or the left valve larger than the 
right, inner duplicatures very broad. Natatory setae of the antennae poorly 
developed. Caudal rami armed with three clavv^s increasing in length distally. 

Males unknown. 

Six species occur in New South Wales. 

Key to species of Ilyodromus. 

A. Surface marked by distinct longitudinal lines. 

B . Ventral margins of the valves almost straight Stanley anus. 

BB. Ventral margins deeply sinuated. 

C. Greatest height not nearly equal to half length. Length exceeding 

L5 mm varrovilUus. 

CC. Greatest height alm.ost equal to half the length. Length less than 

1.5 mm ohtiisus. 

AA. Surface smooth or delicately striated. 

B . Surface quite smooth. . . . . viridulus. 

BB. Surface striated. 

C. Dorsal margin forming an angle above the eye. suhstriatus. 

CC. Dorsal margin evenly curved ellipticus. 

Ilyodromus stanleyanus (King). 

Described by King in 1855 (p. 66) as Candona stanleyana. In 1886 (p. 89) 
Brady described Cypris stanleyana from an immature specimen; it is doubtful 
whether this is identical with King's species. Sars figured this species in 1894 
(Plate v., figs. 3a-e) and transferred it to the genus Ilyodromus. 

Female. Shell, seen laterally, elongated reniform in shape; dorsal margin 



282 FRESHWATER ENTOMOSTRACA OF N.S.W., iii., OSTRACODA, 

straight in the middle, sloping at each end, ventral slightly sinuated in the 
middle; anterior and posterior margins rounded and equal in height. Seen dor- 
sally, very compressed, more pointed anteriorly than posteriorly. Left valve 
very slightly larger than the right, overlapping at each end and ventrally; inner 
duplieatures very broad. Surface of the shell sculptured with elevated longi- 
tudinal lines, with small scattered pits between them. Caudal rami strongly built, 
of uniform breadth; claws well developed. Colour dark green. Length, 1.6 mm. 
Bistrihution. — N.S.W. : Coogee, Maroubra; South Australia; New Zealand. 

IlyODROMUS VARROVIIiLIUS (King). 

Cypris varrovillius King, 1855 (p. 41) ; Ilyodromus varrovillius Sars, 1894 
(p. 41, Plate vi., figs. la-c). 

Female. Shell, seen laterally, narrow and somewhat oblong in shape; dorsal 
margin straight for the greater part of its length, sloping anteriorly and pos- 
teriorly; ventral margin sinuated in front of the middle; posterior rounded, 
lower than the rounded anterior margin. Seen dorsally, moderately compressed, 
oval in outline. Valves and surface of the shell as in the preceding species. 
Caudal rami coarsely built, expanded at the tips, claws well developed. Colour 
deep gi'een. Length, 1.6 mm. 

Distribution. — N.S.W. : Bourke Street, Varroville, Kendall, Holbrook; New 
Zealand. 

Ilyodromus obtusus Sars. 

Sars, 1894, p. 46, Plate vi., figs. 4a-d. 

Female. Shell, seen laterally, reniform, with the greatest height almost at- 
taining half the length so that this species is comparatively higher than the two 
preceding species; dorsal margin very straight, sloping at each end; ventral 
margin deeply sinuated at the middle; extremities obtusely rounded and of equal 
height. Seen dorsally, oval in outline, moderately compressed. Surface and re- 
lative size of the valves as in I. stanleyanus. Caudal rami strongly built, dilated 
at the tips, claws well developed. Colour dark green. Length, 1.4 mm. 

Bistrihution. — This species has not before been recorded in Australia. 
Several specimens were obtained at Leura on the Blue Mountains. Sars' speci- 
mens were reared from dried mud collected at Dunedin, New Zealand. 

Ilyodromus viridulus (Brady). 

Briefly described as Cypris viridula by Brady in 1886 (p. 88). Fully 
described and figured by Sars in 1896 (Plate ii., figs. 3, 4; Plate v., figs. 8-11) as 
Herpetocypris viridulm. Transferred to the genus Ilyodromus by Sars in 1896. 

Distribution.— 'N.S.W.: Bourke Street, Sydney, Condong district, Parra- 
matta; Queensland. 

Ilyodromus substriatus Sars. 

Videns. Sels. • Skrifter i. Math. Nat. Klasse., 1894, No. 5 (p. 45, plate vi., 
figs. 3a-c). 

Female. Shell, seen laterally, of irregular reniform shape; dorsal margin 
slightly curved, angular above the eye; ventral margin distinctly sinuated; pos- 
terior margin rounded, very much higher than the anterior. Seen dorsally, 
moderately compressed, elongated oval in sliape. Valves unequal, the left being 
considerably larger than the right and overlapping it at both ends and ventrally; 



BY MARGUERITE HENRY. 283 

jnner duplieatures broader anteriorly than posteriorly. Surface of the shell 
marked by delicate longitudinal lines, not nearly so conspicuous as in I. stan- 
leyanus. Caudal rami well developed, claws short and strong. Colour light 
yellowish-green. Length, 1.5 mm. 

Distribution.— 1:^.^M^.: Sydnej^, Botany;. New Zealand. 

Ilyodromus ellipticus Sars. 

Sars, 1896 (Plate vii., figs. 4a-c). 

Female. Shell, seen laterally, elliptical; dorsal margin very slightly arched; 
ventral almost straight; anterior and posterior margins rounded and of equal 
height. Seen dorsally, very much compressed, elongated oval in shape with the 
anterior end more pointed than the posterior. Valves almost equal, the left very 
little larger than the right. Surface faintly striated; ends hairy, the hairs be- 
ing long and far apart on the posterior margin. Caudal rami with well developed 
claws. Colour yellow tinged with green. Length, 1.1 mm. on an average, speci- 
mens collected at Lane Cove attained a length of 1.5 mm. 

Distribution. — N.S.W. : Bourke Street, Botany; Lane Cove. 

Subfamily CANDONINAE. 

Antennae composed of five segments in the female and usually six in the 
male. Natatory setae absent. Terminal segment of the second leg bearing three 
unlike setae, two of which are baekwardly directed. The members of this family 
have lost the power of swimming and are found creeping in the mud or on 
water plants. Two genera are represented in New Soiith Wales. 

Key to genera of Candoninae. 

A. Furca normal .. Candona. 

AA. Furca without a dorsal seta . • Candonopsis. 

Genus Candona Baird, 1850. 

Antennae of the female composed of five segments, of the male six. Second 
pair of legs five, sometimes six, segmented, with two baekwardly directed un- 
equal setae and one long forwardly directed seta. Furca normal. Males 
numerous. 

Only one species of this large genus is known to occur in New South Wales. 

Candona lutea King. 

King, 1855, p. 67; Brady, 1886, p. 92, Plate x., figs. 7, 8; PI. viii., figs. 
10, 11. 

Female. Shell seen laterally, subreniform, greatest height occurring behind 
the middle and equalling half the length; dorsal margin moderately arched, slop- 
ing more abruptly posteriorly than anteriorly; ventral distinctly sinuated in the 
middle; anterior end rounded, posterior much higher and obliquely rounded. 
Seen dorsally, compressed, elongated oval in outline, the width equalling one- 
third of the length, anterior end more pointed than the posterior. Valves equal. 
Surface of the shell smooth and polished. Colour dark green; dried shells, 
glistening white. Length, 1.3 mm. 

King's description of this species is very brief and the only mention of the 



284 FRESHWATER ENTOMOSTRACA OF H.S.W.^ Hi., OSTRACODA^ 

appendages is the statement that the antennae lack natatory setae. Brady's 
description is based on dried shells and unfortunately these were the only speci- 
mens available in the present case. 

Distribution. — N.S.W. : Sydney Cove, Condong district, Tweed River. 

Genus Candonopsis Vavra, 1891. 

Antennae as in Candona. Mandible with a very long palp. Furea slender 
and without a dorsal seta. Ten species have been described; one occurs in New 
South Wales. 

Candonopsis tenuis (Brady). 

Candona tenuis, Brady, Proc. Zool. Soc, 1886, p. 92. — Candonopsis tenuis, 
Sars, 1896 (PI. vii., figs. 6a-d). 

Female. Shell, seen laterally, reniform, the greatest height equalling half 
the length and occurring behind the middle; dorsal margin fairly straight, ventral 
sinuated; anterior margin lower than the posterior. iSeen dorsally, very com- 
pressed, anterior end more pointed than the posterior. Valves equal, inner 
duplieatures broad, ends bearing delicate hairs. Surface smooth and polished, 
minutely reticulated. Caudal rami narrow, dorsal seta absent, each terminal claw 
with a denticle situated at half its length. Colour white. Length, .90 mm. 

Male. Dorsal margin arched, ventral sinus occurring in front of the middle, 
anterior end more produced than in the female. Length, 1.0 mm. 

Distribution. — N.S.W. : Bourke Street, Maroubra, Tweed River; Sumatra. 

Family CYTHERIDAE. 

Surface of thie sheU usually rough and uneven. Antennules composed of 
5-7 segments, not adapted for swimming. Antennae composed of 4-5 segments, 
the first of which bears a fiagellum. Natatory setae absent. Three pairs of legs 
which are very alike but vary in size. Furca rudimentary, represented by two 
rounded lobes bearing one or more setae. 

This family comprises few freshwater forms. None have hitherto been re- 
corded from Australia. 

Genus LiMNiCYTHERE Brady, 1868. 

Shell usually thin, tuberculate or spiny. Antennules five-segmented, an- 
tennae four-segmented. Branchial plate of the mandible strongly developed. 
Furea very rudimentary. 

Ten species have been described; a new one is here added from New South 
Wales. 

LiMNiCYTiHERE ASPERA, n.sp. (Plate xxix., figs. 4-8.) 

Female (fig. 4). Seen laterally, somewhat rectangular in shape, higher an- 
teriorly than posteriorly, the greatest height occurring in front of the middle 
about the region of the eye. Dorsal margin straight for the greater part of its 
length, sloping down towards the posterior end and anteriorly forming an arch 
over the eye; ventral margin sinuated in front of the middle; anterior margin 
almost straight, curving slightly to meet the dorsal and ventral margins; pos- 
terior margin curved. Valves almost equal, each marked by a distinct lateral 
furrow. Surface not marked by any definite sculpturing but decidedly rough. 
Antennule (fig. 6) five-segmented, the terminal segment bearing three setae of 



BY MARGUERITE HENRY. 



285 



etjual length and a I'ourtb longer seta. Aiitenna (fig. 7) composed of four seg- 
ments, flagellum lung, unsegmented. Caudal rami (fig. 8) reduced to two 
rounded prominences, each bearing a terminal spine and a very short lateral 
spine. Colour brown. Length, . 41 mm. ; height, . 22 mm. 
Distribution. — N.S.W, : Byron Bay. 



List of icorks referred to. 

Baibd, W., 1859. — Description of some new recent Entomostraca from Nagpur 
collected by Eev. S. Hislope. Proc. Zool. Soc. London. 

Brady^ G. S., 188G. — Notes ou Freshwater Entomostraca from South Australia. 
Proc. Zool. Soc. London, p. 82. 

■ • , 188G. — Notes on Entomostraca collected by Mr. A. Haly in 

Ceylon. Journ. Linn. Soc. Lond. Zool., xix., p. 293. 

^^ — — , 1906. — Notes on the Entomostraeau fauna of the New Zealand 

Lakes. Proc. Zool. Soc. London. Vol. ii. 

Hexry^ M., 1919. — On some Australian Fresliwater Copepoda and Ostracoda. 
Journ. Proc. Roy. Soc. N.S.W. , liii., p. 29. 

King, R. L., 1855. — On Australian Entomostracans. Proe. Potj. Soc. Van Die- 
men's Land, iii.. Part 1, p. 56. 

MuLLER^ G. W., 1898. — ^^Vissenschaftliche Ergebnisse der Reiseu in Madagaskar 
und Ostafrika. Die Ostracoden. Abliand. Senck. Naturf. Ges., 
xxi.. Part 2. 

Ramdohr, F. a., 1808. — Uber die Gattung Cypris Miill. und drei zu derselben 
gelrorige neue Arten. Mag. d. Ges. Naturf. Freunde z-u Ber- 
lin, ii. Jahrg, p. 83. 

Sars, G. 0., 1889. — On some Freshwater Ostracoda and Copepoda raised from 
dried Australian mud. Christ. Videns.-Sels. Forhand. No. 8. 

■ — — , 1894. — Couti-ibutious to the knowledge of the Freshwater Ento- 
mostraca of New Zealand. Videns.-Sels. Skrifter. i. Math. 
Nat. Klasse, No. 5. 

~ , 1896. — On Freshwater Entomostraca from the neighbourhood of 

Sydney. Arch. Math, og Naturvid., Bd. 18, Heft 2. 

■ — , 1896. — On some West Australian Entomostraca raised from dried 

sand. Archiv. f. Math, og Nat. 

Sharfe, R. W., 1903. — Report on the Freslnvater Ostracoda of the United States 
National Museum. Proc. U.S. Nat. Miis., xxvi., No. 1347, p. 
968. 

Zaddach, E. G., 1844. — Syno]>seos Crustacoorum Prussicorum Prodromus. ,: 




L i S 8^ A i^ 



286 FRESHWATER EXTOMOSTEACA OF N.S.W.^ iii., OSTRACODA. 

EXPLANATION OF PLATES XXIV.-XXIX. 

Plate xxiv. 

Newnhamia fenestrata. 

Fig. L— $. Lateral view (x 60) ; Fig. 2.— dorsal view (x 60) ; Fig. 3.— surface 
markings (x 225); Fig. 4. — shell margin (x 225); Fig. 5.— ventral plate (x 60); 
Fig. 6.— end segment, second leg (x 225); Fig. 7. — furca (x 358); Fig. 8.— end of 
antenna (x 225) ; Fig. 9.— second maxilla (x 550) ; Fig. 10. — second maxilla d". (x 
550). 

Plate XXV. 

Cypridopsis australis, ?. . 

Fig. 1. — Lateral view (x 88); Fig. 2. — dorsal view (x 88); Fig. 3. — surface 
(x 200); Fig. 4.— antenna (x 408); Fig. 5.— first leg (x 550); Fig. 6.— second leg 
(x 550) ; Fig. 7.— furca (x 350) . 

Plate xxvi. 

Cypris crmita. 

Fig. 1.—?. Lateral view (x 25); Fig. 2.—^. lateral view (x 25); Fig. 3.—?. 
dorsal view (x 25); Fig. 4. — $. antennule (x 62); Fig. 5. — ?. antenna (x 62); 
Fig. 6.—$. first leg (x 70); Fig. 7.—?. second leg (x 70); Fig. 8.—?. furca 
(xoO). 

Plate xxvii. 

Cyprinotus tenuis, ?. 

Fig. 1. — Lateral view (x 67); Fig. 2. — dorsal view (x 67); Fig. 3. — surface 
(x 270); Fig. 4.— antenna (x 166); Fig. 5.— antennule (x 166); Fig. 6.— first leg 
(x 166); Fig. 7.— second leg (x 166); Fig. 8.— furca (x 170). 

Plate xxviii. 
Figs. 1-5. Cypretta hirsuta, ?. 

Fig. 1. — Lateral view (x 105); Fig. 2. — dorsal view (x 105); Fig. 3. — valve- 
margin (x 210); Fig. 4. — end segment, second leg (x 525); Fig. 5. — furca (x 525). 

Figs. 6-7. Ilerpetocypris laevissima, ?. 

Fig. G.— Lateral view (x 72); Fig. 7.— furca (x 100). 

Plate xxix. 

Figs. 1-3. Cyclocypris tenuissima. 

Fig. 1.— c?. Lateral view (x 94); Fig. 2.—?. lateral view (x 94); Fig. 3.— 
$. furca (x 468]. 

Figs. 4-8. Limnicytliere aspera, ?. 

Fig. 4. — Lateral view (x 94); Fig. 5. — first leg (x 450); Fig. 6. — antennule 
(x 450); Fig. 7.— antenna (x 450); Fig. 8.— furca (x 450). 



287 



ANATOMICAL FEATURES OF THE MATURE SPOROPHYTE OF 
SELAGINELLA ULIGINOSA. 

By Jessie K. Steel^ B.Sc. 

(With nineteen Text-figures.) 

{Communicated by Professor A. A. Lawson.) 

[Read 25th July. 1923,] 



Introduction. 



A classification of vascular plants, based on theii' reproductive and anato- 
mical features, has been made by Professor Jeflirey (1897). According to this 
classification, vascular plants may be divided into two large groups. The first 
group, the Lycopsida, includes the lycopods and their allies, and is characterised 
b^' plants bearing small leaves and ventral or adaxial sporangia. The vascular 
cylinder xaa,j be either siphonostelic or polystelic, but is distinguished by the 
fact that no foliar gaps occur. On the other hand, in the second group, the 
Pteropsida, the plants have relatively large leaves, and foliar gaps are a very 
characteristic feature of the vascular structure. The sporangia in these types are 
abaxial in position and, unlike those of the Lycopsida, are numerous. In the 
higher members of the group true seeds are developed. The Pteropsida include 
the ferns, and the higher vascular plants, namely the Gymnosperms and x\ngio- 
S]3erms. 

Dr. Scott, however, is of the opinion that this classification should be slightl}'' 
altered. Pie claims that the Lycopsida should comprise the Lycopodiales only^ 
while their allies, such as the Equisetales, Sphenophyllales and Psilotales, should 
be placed in a separate group, to Avhich he gives the name Sphenopsida. The 
affinities existing among the members of this last grouiD and the Lycopods are not, 
in Scott's opinion, sufficient to enable them to be classified together. The term 
Pteropsida is still retained by Dr. Scott to include the Filicales, Oymnosperms 
and Angiosperms. 

Of the more important of the earlier investigations made on the anatomy of 
the mature sporophyte of Selaginella, mention may be made of the work of 
Spring (1849) who draws attention to the difference in anatomy in the creeping 
and erect stem. Later we come to the work of De Barj', who described the 
vascular bundle as being of the true fern type. Subsequent writers, as, for 
example, Vladescue (1889), were occupied with an investigation of the nature of 
the layers immediately surroi;nding the vascular cylinder — the endoderrais and 
perieycle. 

Towards the close of the nineteenth century, we have contributions to our 



288 THE MATURE SPOROPHYTE OF SELAGINELLA ULIGIKOSA^ 

knowledge of the geuus made by Strassburger and Van Tieg'hem. It is, how- 
ever, to Professor E. J. Harvey Gibson that we awe most of our knowledge of 
the anatomical features of Selaginella as a whole. In a series of articles (1894- 
1902) he has given us a verj'^ exhaustive account of the anatomy of the stem, 
leaf, ligule and root of a great number of species. Harvey Gribson's work was 
further carried on by Miss Mitchell (1910), who undertook an investigation of 
the structure of the strobilus. The nature of the root and the rhizophore, which 
is present in some species, has also been studied by Uphof (1920). 

It is vv^ith a view to advance our understanding of the local species of 
Selaginella, that the following investigations have been made. These have been 
carried out entirely on the mature sporophyte, no attention being paid to cyto- 
logical details in connection with the gametophytes or young embryos. 

The Plant. 

Selaginella uliginosa is common to Eastern Austi'alia, and grows proliflcally 
both in very moist and in dry soil. The gTeater part of the material for this 
study was collected on the cliffs at Coogee, near Sydney, New South Wales, where 
it was growing under extremely moist conditions. Other material, however, both 
from the National Park and from the Blue Mountains, New South Wales, was 
growing equalh" well in exactly reverse conditions. 

This species of Selsginella presents an interesting periodicity in the develop- 
ment of its vegetative organs. The aerial parts of the plant die away, generally 
after a period of reproducti^"e activity, and .fresh shoots arise from the under- 
ground rhizome, enabling the plant to carry on its activities for a j^rolonged 
season. 

External, Morphology. 

In its external morphology, this species of Selaginella somewhat resembles 
S. laevigata as described by Harvey Gibson, but the internal anatomy of the two 
forms shows many distinguishing features. S. uliginosa is characterised by a very 
distinct rhizome (Text-fig. 1) which may sometimes branch, and which varies 
from about .2 to .tt cm. in diameter. The rhizome bears erect aerial shoots, with 
leaf bases in the lower portions. These shoots are found in close jDroximity to 
the point at which the root is given off, a feature characteristic of the Lycopsida 
in general. They are about .1 cm. in diameter in their thickest part, and grow 
to a lieight of 10 to 12 cm., although some were fou.nd of much greater length. 
The slioots bear two rows of lateral branches, bearing the leaves Avhich are A^ery 
small, acut^e, sessile and decussate. 

Tlic reproductive organs are found in square shaped cones, either terminating 
tlie main axis or the lateral branches. Numerous specimens, however, were found 
in wliich vegetative growth had continued from the axis of the cone, after the 
completion of its reproductive activity (Text-fig. 2). Attention has already been 
drawn to this fact by Osborn (1915). The occurrence of alternating sterile 
and fertile regions is suggestive of the conditions found in tlie more i)rimitive 
Lycopodiales. This, togethei- with other facts relating to the distri})ution of the 
.sporangia, to be described )at(M', v/ould tend to place S. uliginosa among the lower 
species of the genus, since it shows close affinities to such types as Lycopoclium 
tela go. 

S. uliginosa is diaracterised by the absence nl" tlie so-called rhizophores, a 
l)oint in which it resembles .S', cuspidnta and S. hwrigata. The roots which, as 



CY JESSIE K. STEEL. 



289 



al/eady mentioued, axise ir,. close proximity to the aerial shoots, average .5 to 1 
mm. in diameter, and may give off lateral roots of much smaller dimensions. 



Anatomy. 

In describing the microscopical anatomy of this type, the writer has, for con- 
venience, adopted the nomenclature for the various tissues given by Harvey 
Gibson in his work on I he anatomy of the genus. These terms, as he points out, 




Text-tig. 1. — Habit Study of S. uliginosa. 

Text-fig. 2. — Aerial slioot of S. uliginosa showing alternating reproductive and 
sterile regions. 

are used purely "in a descriptive sense." The limiting layer of the stem, he has 
called the "epidermis," while the sclerised layer, usually lying inside this, is known 



290 



THE MATURE SPOROPHYTE OP SELAGIKELLA ULIGIXOSA, 



as the "liypodermis." This gradually merges into the ordinary parenchymatous 
cells of the "cortex." The unicellular or multicellular strands uniting the vascular 
cords to the cortex are termed ''trabecular"; these may be composed of either an 






Text-fig. 3. — T. S. Aerial stem S. uliginosa. a, epidermis; h, cuticle; c, hypo- 
dermis; d, cortex; e, trabeculae; /, steles, (x 40). 

Text-fig. 4. — T. S. Stoles taken at different levels in tlie aerial stem of S. 
uliginosa. (x 40). 

Text-fig. 5. — T. S. Single stele S. tilifjinosa. a, endodermis; />, perieyele; c, 
phloem: <•/, protf)xylem; e, primary xylem; /,, trabeculae. (x 320). 



endodenual cell or two or more parencliymatous cells, connected with the endo- 
dermal cell to form a trabecula. Tlic ''endodemais" is compf)sed of cuticularised 
cell.s arising from the clilorophyllaccons layer surrounding the stele. This layer 



BT JESSIE Iv. STEEL. 291 

Harvey Gibson has called the ''pericycle." The term "vascular bundle" is used to 
indicate a leaf-trace only, while the "stele," refers to the vascular strand, enclosed 
within the so-called endodermis and perieycle. 

The Aerial Stem. 

In transverse section, the aerial stem of 8. uliginosa (Text-fig. 3) is seen to 
have a distinct epidermis, with a very well marked cuticle. In sections towards 
the base of the shoot, there is a well defined hypodermis of sclerenchymatous cells 
gradually merging into the thin-walled cells of the cortex. This hypodermis is 
r^ot so well marked in sections near the apex. The centre of the section is oc- 
cupied by a large lacuna, traversed hj unicellular or multicellular trabeculae which 
serve to connect the stele or steles, as the case may be, to the cortex. 

Sections were cut at various positions in the aerial stem, and up to as many 
as four steles were found (Text-fig. 4). In some of the younger shoots examined, 
however, this division into four had not taken place. As the apex of the shoot is 
reached, the four steles gradually become joined, until finalljr only one remains. 

On closer examination of the stele, it was seen to have a well-defined endo- 
dermis (Text-fig. 5) of relatively large cells, with a perieycle, the radial walls of 
each cell of which appeared thickened. The phloem is composed of protophloem 
and phloem parenchyma. It completely surrounds the xylem, but is poorly 
developed in the region of the protoxylem. In sections shoAving four steles, the 
protoxylem of each stele lies to the outside of the primary xylem. As the steles 
join up, this position is retained, until finallj^, when one stele remains, it is seen 
to have a gi'oup of protoxylem at either side. The tracheids of the xylem show 
well-marked spiral and scalariform thickening. It must be noted, in connection 
with the vascular structure of the aerial stem, that at the point where the vascular 
bundle goes to the lea,ves, no gap occurs. This is one of the chief characteristics 
which place the Selaginellas among the Lycopsida. 

The Leaf. 

The leaf, in transverse sections (Text-fig. 6), showed a thick cuticle developed 
on both upper and lower surfaces. The epidermal cells are fairly regular in 
shape; stomata each possessing two guard-cells occur on the lower surface of the 
leaf (Text-fig. 7) and are developed in the region of the midrib. 

There is no development of palisade tissue in the leaf of *S'. uliginosa, the 
bulk of the leaf being composed of a very loose network of mesophyll cells (Text- 
fig. 7) containing chlorophyll and with abundant air spaces between. In the 
region of the midrib, the cells are very much enlarged and rather mucilaginous, 
possibly serving to retain water. Two ' groups of someAvhat thick-walled cells 
occur on either side of the vascular bundle (Text-fig. 8). These are young 
tracheidal cells and in some a definite thickening could be seen on the walls. No 
definite endodermis and pei^cyele were observed in the leaf, except perhaps at the 
extreme base, where the vasevilar bundle first arises from the stele of the aerial 
stem. The vascular bundle is composed of xjdem and phloem, the former lying 
to the upper side of the leaf, and composed of the characteristic scalariform 
tracheids. 

The Ligule. 

In close association with the leaves of S. uliginosa is found the ligule. It 



292 



THE MATURE SPOROPHYTE OF SELAGINELLA ULIGIKOSA_, 



occurs in the axil ])etween the upper surface of tlie leaf and the aerial shoot. 
For the most part, the ligules occurring with the sporophylls of the cone seemed 
much better developed; several good examples were found, however, in association 
with the leaves. In longitudinal section (Text- tig. 9) the vascular bundle running 
out from the central stele to the leaf, was seen to be in close promixity to the 
base of the ligule. 

As has been described by Harvey Gibson, the ligule is composed of a masa 




Text-tig. 6. — T. S. Leaf ,S'. uliyinosa. a, cuticle; h, epidermis; c, mucilaginous 
cells; d, group of thick-walled cells; e, young tracheidal cell; 
/', vasculai- bundle, (x 150). 

Text-fig. 7. — t. S. Lower epidermis of leaf of S. idlpinoi^a. a, stomata; &, 
guard-cells; c, mesophyll. (x 310). 

Text-fig. 8. — T. S. Vascular bundle of leaf of ^S*. iiUg'mosa. a, young trachei- 
dal cells; h, mucilaginous cells; c, xylem; d, phloem, (x 310). 

Text-fig. 9. — L. S. Ligule H. tilifp'nosa. a, glossopodium ; /^ sheath; c, vascular 
liuiidlc rnniiiiii:' to leaf, (x 310). 



of cells witli ratluT dense protoplasmic contents. It is swollen ;it tlie bsiso, form- 
ing the glossopodium. composed of very hirge thin-walled cells, and surrounded 
by a sheath of narrow cells. ]n,S'. uUffhosa thx^ ligule tapers voiy much towai'ds 
the ai»cx. the cells possessing dense granulai' contents a7id v<tv distinet nuclei. 



BY JESSIE K. STEEL. 



293 



The Rhizome. 

As has been previously mentioned, the rhizome forms a very characteristic 
feature of S. ulic/iuosa. In transverse section (Text-fig. 10) tlie epidermis shows 
a thick cuticle, its cells being square in shape. Within the epidermis we find a 
relatively wide cortex, composed towards the outside of a hypodermis of sclerised 
cells which gradually merge into the smaller and thin-walled cells of the inner 
cortex. 

The rhizome of S. uliginosa is solenostelic in its vascular structure, but the 




Text-fig. 10. — T. S. Rhizome S. uliginosa. a, cuticle; &, epidermis; c, hypo- 
dermis; cl, cortex; e, phloem; /, protoxvlera; g, primary- xvlem; 
h, pith, (x 125). 

Text-fig. 11. — T. S. vascular cylinder of rhizome of S. uliginosa^ showing branch 
gap. a, cortex; h, phloem; c, x^'lem; d, pith, (x 70). 

Text-fig. 12. — T. S. root 8. uliginosa. a, epidermal cells; h, hypodermis; c, 
cortex; d, stele, (x 70). 

Text-fig. 13.- — T. S. Stele of root of S. uliginosa. a, cortex; h, thick-walled en- 
dodermal cell; c, thin-walled endodermal cell; d, phloem; e, pro- 
tnxylem; /, jirimary xylem. (x 310). 



294 THE MATURE SPOROPHYTE OF SELAGINELLA ULIGIKOSA^ 

endoclei-mis and pericycle are not quite as definite as in the aerial stem. The 
greater part of the vascular cylinder is made up of the scalariform tracheids of the 
xylem, with internal and external phloem. The centre of the section is occupied 
by the thin-walled cells of the pith. 

The solenostelic character of the rhizome of S. uliginosa is of particular in- 
terest phylogenetieally. This point is emphasised by Tansley, who says "The 
solenostelic or siphonostelic vascular cylinder seems an intermediate step between 
the primitive protostelie types on the one hand, and the more highly advanced 
dialystelic types on the other." Bower also refers to the occurrence of this con- 
dition, in describing the anatomy of the axis of Selag'mella, in the following 
words: "In the more complex cases, the axis becomes polystelic or in some eases 
solenostelic (rhizome of 8. laevigata) thus resembling similar vascular complica- 
tions seen in the stems of Ferns. These may be held to be relatively late, and 
special developments from the non-medullated, monostelie type; their origin shows 
parallelism of development rather than any nearer relation with the similar struc- 
ture seen in the Ferns." 

At the point in the rhizome from which a branch is given off, a distinct gap 
occurs in the vascular cylinder. A transverse section made at this point (Text- 
fig. 11) shows the phloem continuous round the end of the xylem, uniting with 
that of the inner side. At the gap, the parenchymatous cells of the cortex come 
m direct contact with those of the pith. It is important to note that in this case, 
although the gap is prolonged forward, still no overlapping takes place, a step 
which would lead to a dietyostelic condition. The occun-ence of the gap, however, 
has an interesting bearing on the relationship of this type with higher forms. 

The Root. 

The root exhibits several characteristic features. In the first place, the 
epidermis is not so clearly differentiated from the underlying eeUs as in the other 
organs examined, and the cuticle, although present, does not show the extensive 
thickness met with in other sections (Text-fig. 12). In several sections of the 
root, there appear several rows of parenchymatous ceUs within the epidermis, 
while in others it gives place directly to a thick-walled hypodermis, an outstanding 
feature of the root section. Within this hypodermis lies a relatively wide cortex 
of parenchyma enclosing a single stele. 

Unlike the rhizome, the root possesses a well defined endodermis and peri- 
cj'cle. The endodermis is usually one layer in thickness (Text-fig. 13) and is 
composed, for the most part, of sclerised cells. In the places where this thicken- 
ing has not taken place, the endodermal cells can still be traced. The pericycle 
is composed of parenchymatous cells, within which lie the elements of the phloem. 
In the root, the primary xylem is usually made up of three or four very large 
tracheids, there })eing a single group of protoxylem elements at one side. The 
phloem is poorly developed in this region. 

The Apices of tlie Bhisome and Boot. 

The gi'bwiiig region of S. uliginosa lias some interesting points. Growth 
could not be traced to a single apical cell, but rather to a group of cells lying 
just l)elow the dermatogen and possessing very dense cytoplasmic contents. Very 
many sections of the apex of the rhizome were made, but in none was there (!vi- 
(lence of a single apical cell, as in many otlier Ptcridophytes. Instead, the der- 
matogen was continuous over the apex, while lying b^'neatli it was a small group 



BY JESSIE K. STEEL. 



295 



of cells (Text-fig. 14), one or two of wbieh appeared to possess two nuclei, as if 
the cells were in the act of division. Other cells, adjacent to tliese, seemed to ha.ve 
recently divided, although no trace of any spindle could be seen. It therefore 
seems highlj^ probable that growth in 8. uJiginosa takes place from a group of 
meristematic cells and not from a single apical cell. 

Attention has been drawn to the phylogenetic significance of this interesting 
point by Bower (1889) thus: "In many species of Selaginella a single initial cell 




Text-fig. 14. — L. S. Apex of rluzome of *S'. uliginosa. a, meristematic cells. 

(x 710). 
Text-fig. 15. — L. S. Apex of rhizome of S. uliginosa showing enveloping leaves. 

(x 175). 
Text-fig. 16.— L. S. Apex of root of S. uliginosa. a, young traeheidal cells; 

h, root-cap. (x 175). 

is formed in the subaerial stem, in other species there is a more complex structure 
of the meristem, while in Lyeopodium three initials have been observed, though 
the general habit is similar to that of the allied Selaginellas." The complex 
structure of the meristem may therefore be regarded as a primitive feature, since 



296 



THE MATURE SPOROPHYTE OF SELAGINELLA ULIGIXOSA, 



it is found in the genus Lyeopodium and also in S. spimilosa, a type which on 
the grounds of its radial shoot and anatoniioal structure is regarded as essentially 
primitive. 

Another interesting feature of the apex is the very characteristic manner in 
which the young leaves envelope it, as shown in Text-fig. 15. 

In regard to the apex of the root, the dense cytoplasmic nature of the cells 
was again noted. There is a distinct root-cap of large, loose, thin-walled cells. 
Some distance back from the apex, the differentiation into dermatogen, periblem 
and plerome could be seen (Text-fig. 16). Several very large nuclei were found 
in the centre of the section, occurring in elongated cells which would later develop 
into the tracheids of the older root. As in the apex of the rhizome, no definite 




Text-fig. 17. — T. S. Cone of S. uliginosa. a, stele; h^ sporangia; c, spoi-opiiyli. 
(x 160). 

apical cell wa.s observed, so that it is probable that in the root, wc have a meristem 
similar to "tliat described above for tlie rhizome apex. 

Spore Vrddnci'ujn. 

As ali'eady slated, tlie cones of ,S'. iili(jinosa citlior occur at the apex of the 
lateral branches or tciininatc the maiii axis. In transverse sectitm they appear 
almost sfjuare (Text-(ig. 17), tlie analoniical details cbxsely agreeing with those 
of the aei-ial stem. 'I'lic cuticle of the (■|)ideriiiis is again a feature of note, but 
is not so well develoi^'d on llie epidermis bordering the sporangia. The sporo- 
phylls ai-e arranged in four sets around the central axis, the vascular structure 
of this axis recalling Hiat of Ihc vpgctative shoot. Tn the centre is a large lacuna, 



BY JESSIE K. STEEL. 



297 



traversed by unicellular or multicellnlfu' tra,beculae whicli hold the single stele of 
the axis in position. The endoderaiis and perieyele are again clearly differen- 
tiated; the phloem surrounds the main mass of xylem which possesses two groups 
of protoxylem. The sporophyll bundles join the axial strand in the region of the 
protoxylem, the endodermis being continuous for a very short distance at the 
base of the sporophyll. Oblique sections of these bundles are often found in the 
cortex of the central axis. They are essentially similar in their vascular anatomy 
to the ordinary vegetative leaves. 

In longitudinal section (Text-fig. 18) some further interesting anatomical 
features are to be noted. The single central stele of the axis is very conspicuous, 




Text-fig. 18, — Sketch of L. S. Cone of S. uliginosa. a, sporophyll; li, flap; c, 
stele of axis; d, megasioorangium; e, microsporangium. 

Text-fig. 19. — L. S. Megasporangium of S. uliginosa. a, line of dehiscence; h, 
wall; c, megaspore; d^ stalk, (x 160). 



and from this vascular bundles are given of£ to the sporophylls. The latter are 
prolonged into a distinct flap on the lower side, and bear sporangia on the upper. 
The occurrence of this well developed dorsal flap protecting the sporangia in many 
species of Selaginella is regarded by Sykes and Styles as a primitive feature, re- 
calling the condition found in such types as Lepidostrobus and Spencerites. The 
sporangia of S. ulig-inosa are borne in the axils of the sporophylls, so that they 



298 THE MATURE SPOROPHYTE OF SELAGINELLA ULIGIXOSA^ 

arise from stem tissue and not from leaf tissue as in the more primitive members 
of the Lyeopodiales. 

Lying between the sporangia and the sporophyll is the ligiile. This shows 
the same anatomical features as have been i:)reviously described for this organ as 
it occurs with the vegetative leaf, but it is more perfectly developed. The vascular 
bundle of the sporophyll is in close proximity to the base of the ligule. but no 
bundle is developed to the sporangium. 

As regards the distribution of the microsporangia and megasporangia, S. 
ulighiosa shows some interesting features. On examining the very young cones, 
the writer found that only microsporangia were present, but later, however, mega- 
sporangia make their appearance. Numerous examples were found in which the 
megasporangia were borne towards the centre of the cone, while microsporangia 
were lying both above and below. 

A similar arrangement to this has been noted by Sykes and Styles in S. pumila 
and Miss Mitchell also finds this indiscriminate arrangement in such types, as 8. 
Martensii, S. eanlescens, S. patula, S. cusptdata and others. Phylogenetically, this 
feature is of great importance, and must be regarded as distinctly primitive, the 
more advanced types being those in which the microsporangia are borne towards 
the apex while the megasporangia lie at the base of the cone, giving a possibility 
of pollination by gravity. 

In a few eases of the cones of S. uUginosa examined, an abortive sporangium 
occurred at the apex of the cone, being a possible indication that a second vege- 
tative phase was being entered upon. No cones were found showing megaspor- 
angia only. 

In structure, both sporangia show a very much thickened wall. The micro- 
sporangia all contain numerous microspores, with a distinct tetrahedral division 
and a relatively smooth wall. On the other hand, the megasporangia { Text-fig. 
19) in all cases examined, were found to contain four large megaspores, giving a 
four-lobed appearance to the sporangium. Each megaspore possesses a thick 
spiny wall. As has been previously stated, no examination of the gametophytic 
ceUs was made, as the material had not been preserved with a view to the eyto- 
logical examination of these features. 

Summary. 

The chief facts to be gathered from the above investigation are as follows : — 

1. S. till gin osa is characterised by the presence of a radial type of shoot and a 
distinct rhizome. The aerial shoots and roots are given off from the rhizome 
in close proximity to one another. 

2. The rhizome is solenostelic, with gaps at the point of departure of a branch. 
.3. The vascular structure of the aerial stem shows as many as four steles. 

4. The leaves are small, decussate, and well adapted to xerophytio conditions be- 
cause of their thick cuticle, relatively few stomata, and the occurrence of 
mucilaginous and tracheidal cells in the region of the vascular bundles. 

5. The ror.t is protostelic ; there is only one group of protoxylem. 

G. Growt]i, both in rhizome and root, pi-obably takes place from a gr-oup of 
meri.stematic cells lying below the derinatugen, rather than from a single apical 
cell. 

7. The cones are borne at tlie apices of the main branch or lermitiating the 
lateral shoots. A "selago" condition is common. 



BY JESSIE K. STEEL. 299 

8. Microsporangia and megasporangia shoAv an indiscriminate arrangement, the 
latter, when present, generally occupying the centre of the cone. 

9. The microspores are numerous, but there are only four large megaspores^. 

The above facts lead us to conclude that Selaginella uUffinosa may be classed 
among the more primitive species of the genus. In relation to its external mor- 
phology, we may quote the words of Bower, "The discussion of tlie external mor- 
phology of the genus Selaginella has led to the recognition of the radial type as 
relativel}^ primitive, Avhile those species with dorsiventral shoots are held to be 
more specialised and derivative." This, together with the frequent occurrence of 
a Selago condition, the mixed arrangement of the sporangia in the cones, and the 
presence of four megaspores within the megasporangium, all i^oints to a close 
relationship Avith the more primitive members of the IJycopodiales. On the other 
hand, in the anatomy' of the rhizome, there appears to be a slight advance on the 
primitive condition. It must be remembered, however, that the gaps formed in 
the solenostele are caused by the departure of a branch, and not by a leaf trace 
as in the more advanced Pteridophytes. 

A study of such primitive types as S. uliginosa must lead to a clearer know- 
ledge of the importance of the genus among other vascular plants. Its hetero- 
sporous character, combined with its relatively simple vegetative anatomy, makes 
Selaginella one of the most interesting types, and one deserving very detailed 
examination. 

In conclusion, the writer wishes to express her indebtedness to Professor A. 
A. Lawson and Dr. J. McLuckie for the advice and help they have given her in 
the more difficult problems connected with this study. 

List of references. 

Bary, Antox de, 1884. — Comparative Anatomy of the Vegetative Organs of the 

Phanerogams and Ferns. Translated by F. 0. Bower and D. 

H. Scott. 
Bower, F. O., 1889. — The Comparative Examination of the Medstems of Ferns 

as a Phylogenetic Study. Ann. Bot., iii., 305-392. 
, 1894. — A Theory of the Strobilus in Archegoniate Plants. Ann. 

Bot., viii., 343-365. 

, 1908.— Origin of a Land Flora. 

Campbell, D. H., 1905.— The Structure and Development of Mosses and Ferns. 
Gibson, Harvey, 1894. — Contributions towards a knowledge of the Anatomy of 

the Genus Selaginella (Spr.). Part i. The Stem. Ann. Bot.^ 

viii., 133-206. 

^ 1896.— Id. Part ii. The Ligule. Ann. Bot., x., 77-88. 

,1897.— Id Part iii. The Leaf. Ann. Bot., xl, 123-154:. 

_. ^ 1902.— Id Part iv. The Root. Ami. Bot., xvi., 449-466. 

Jeffrey, E. C, 1897. — The Morphology of the Central Cylinder in Vascular 

Plants. Bep. Brit. Assoc. Adv. Sci., Toronto, 869-870. 
, 1899.— The Morphology of the Central Cylinder in Angiosperms. 

Trans. Canadian Inst., vi. 
'■ , 1902-3.— The Structure and Development of the Stem in the 

Pteridophyta and G;\Tnnosperms. Phil. Trans. Boy. Sac, B. 

Vol. 195, '119-146. 
, 1917.— The Anatomy of Woody Plants. 



300 THE MATURE .SPOROPHYTE OP SELAGINELLA ULIGINOSA. 

Mitchell^ G., 1910. — Coutributions towards a knowledge of the Anatomy of the 

Genus Selaginella. Part v. The Strobilus. Ann. Bot., xxiv., 

19-33. 
OSBORN^ T. G. B., 1915. — ObseiTations on the Morphology of S. uliginosa. Hep. 

Brit. Assoc. Adv. ScL, 727. 
Scott, D. H., 1909.— Studies in Fossil Botany. 
Spring, 1849. — MonogTaphie de la famille des Lycoijodiacees. Pt. ii. Mem. 

Acad. roy. helg. 
Sykes and Styles, 1910.— -The Cones of the Genus Selaginella. Ann. Bot., xxiv., 

523-536. 
Taksley and Chick, 1903. — -On the Structure of Sclvicaea malaccana. Ann. Bot., 

xvii., 493-510. 
Tansley, 1908. — Lectures on the Evolution of the Fiiicinian Vascular System. 

New Phytologist, Reprint No. 2, 1-143. 
Uphop, 1920. — Contributions towards a Knowledge of the Anatomy of the Genus 

Selaginella. Ann. Bot., xxxiv., 493-517. 
Vladescue, 1889. — Communications preliminaires sur la structure de la tige des 

Selaginelles. Journ. de Bot., 261-266. 



301 



STUDIES ON AUSTRALIAN MOLLUSC A. Part xiv. 
By C. Hedlet, F.L.S. 
(Plates xxx.-xxxiii., and one Text-fl^re.) 
{Continued from Vol. xli., p. 719.) 
[Read 25th July, 1923.] 



Arca obtusoides Nyst. 



Area ohtusoides, Nyst, Tabl. syn. Arcacees Mem. Acad. Roy. Belgique, 
xxii., 1847, p. 50 for Arca ohtusa Reeve, Conch. Icon., ii., 1844, PI. xii., fig. 77, 
not Cucullaea ohtusa Phillips, 1836. — Arca ohliquata, Reeve, not Grray, op. cit., 
fig. 80. — Arca sinensis, Philippi, Zeitsch. f. Malak., viii., 1851, p. 53. — Barbatia 
deeurvata, Dunker, Index Moll. Mar. Jap., 1882, p. 232; Lamy, Journ. de Conch., 
Iv., 1907, p. 73. 

Specimens from Moreton Bay were identified by Mr. E. A. Smith in 1889 
as A. sinensis. I gathered a series at the mouth of the Annam River in August, 
1916. 

Berthelizstia. 

BertJielinia, Crosse, Journ. de Conch., xxiii., 1875, p. 39, PL ii., fig. 3; 
Moreh, Journ. de Conch., xxiv., p. 374; Crosse and Fischer, Journ. de Conch., 
XXXV., 1887, p. 305, PI. x. ; Fischer, Man. Conch,, 1886, p. 950; Cossmann, Ann. 
Soc. Roy. Malac. Belg. xxii., 1887, p. 170, PL vii., figs. 24-29, PL viii., figs. 1-2. 

Referring to an article on Edenttellina (Hedley, Proc. Malac. Soc, xiv., 1920, 
pp. 74-76), Mr. M. Cossmann wrote to me, "En effet vos figures dJEdentellina 
sont identiques a celle de BertJielinia Crosse," the "identiques" is doubly under- 
lined. As I have not seen Berthelinia I cannot express an opinion on this 
synonymy. 

This identification lends an unusual interest to the Australian species as 
indicating a remarkable longevity. For Berthelinia was based on a fossil from 
the Eocene of the Paris basin. This eccentric type was first regarded as related 
to Capulus, then Morch transferred it to the Pleurobranchiidae and Fischer 
finally assigned it to the true position in the Juliidae. 

Pedum spondyloideum Gmelin. 

Ostrea spondyloidea, Gmelin, Syst. Nat., xiii., 1791, p. 3335, not Ostrea 
spondylodes, Gmelin, op. cit., p. 3321. — Ostrea pedum, Bolten, Mus. Bolt., 1798, 
p. 170. — Pedum spondyloideum. Reeve, Conch. Icon., xi., 1858, PL i., fig. 1; 
Chenu, Illustr. Conch., 1843, Pedum, p. 1. 

A genus is added to the Australian fauna by a single notched valve of this 
species, which I gathered on the beach of Lizard Island, 19/7/16. 



302 STUDIES ON AUSTRALIAN MOLLUSCA^ xiv., 

Spondylus an acanthus Mawe. 

Spondylus anacanthus, Mawe, Linn. Syst. Conch., 1823, p. 47, PI. xi., fig. 
3; Fulton, Journ. of Conch., xiv., 1915, p. 359. 

This distinct species is as yet unrecorded for Australia. On the beaches of 
Palm and of Lizard Islands I have gathered a few dis-associated valves. 

Lima strangei Sov>^erby. 

Lima strangei, Sowerby, Conch. Icon., xviii., 1872, PI. iii., sp. 15. 

In his recent monogTaph of the family Limidae in the Conchylien Cabinet, 
Dr. J. Thiele follows the Challenger Report in reducing Lima strangei to a 
synonym of L. buUata. The original figure of L. hullata (Born, Mus. Caes. 
Vindobon., 1780, p. 110, PL vi., fig. 8) shows a shell, ascribed to Barbadoes, 
broader and more infiated than our form. Dr. F. Brauer, who re-examined 
Born's type, confirmed (Sitz. k. Akad. Wjss., Ixxvii., 1878, p. 23) the specific 
identity of Sowerby's figure of a Philippine specimen of L. hullata, in the Con- 
ehologiea Iconica 3a, 3b, with Born's type of L. hullata. Certainly there is a 
difference between L. htdlata and L. strangei. 

It appears to me that Lima jacksonensis (Thiele, Conch. Cab., vii., 1920, 
p. 51, PI. 9, fig. 26; type locality. Port Jackson) is actually based on a young 
specimen of L. strangei and thus an Australian species is • simultaneously rejected 
as strangei but restored as jacksonensis. 

To L. jacksonensis Dr. Thiele doubtfully referred the Tasmanian Tertiary 
fossil erroneously identified by Tenison Woods (Proe. Roy. Soc. Tasm., 1876 
(1877), p. 113) as L. subauriculata Montagu. But Tate had already made the 
correction by naming (Proc. Roy. Soc. Tasm., 1884 (1885), p. 230) that fossil 
L jeffreysiana. 

Modiolus areolatus Gould. 

Mytilus (Modiola) areolatus, Gould, Proc. Bost. Soc. Nat. Hist., iii., 1850, 
p. 343 and Moll. Am. Expl. Exped., xii., 1852, p. 452, PI. 41, fig. 562.— Mo d/iolus 
australis, May, Illustr. Index Tas. Shells, 1923, PI. iv., fig. 7. 

There is a large bearded mussel common to New Zealand and Tasmania and 
which reaches north to the latitude of Sydney. This has been generally accepted 
as Modiola australis Gray, a view of its identity expressed by Pritehard and 
Gatliff (Proc. Roy. Soc. Vict., xvii., 1904, p. 251) and by Suter (Man. N.Z. 
Moll., 1913, p. 867). But writers have not been quite unanimous, for early in 
its literary history Menke proposed (Zeits. Malak., i., 1844, p. 63) to unite M. 
australis with M. alhicosta Lamarck, a judgment that was probably correct. The 
few remarks Gray made about his single worn and unlocalised valve are a poor 
foundation for a species. In 1857 Reeve gave a figure of M. australis based on 
material brought from Queensland by Jukes and which may have belonged to a 
different species. This shows a high pitched dorsal margin and very prominent 
umbos, a contour, in short, quite unlike the southern form with which this 
name is usually associated. 

It is now suggested that Reeve's fig-ure of M. australis actually represents 
the shell named Modiola plumescens by Dunker and refigured by Lynge and 
Odhner. But whether Gray's M. australis be alhicosta or plumescens or neither, 
it is incompatible with the southern shell. 

So it is proposed to advance for service in place of the misused M. australis, 
Gould's name of M. areolatus which is clearly identifiable from good figures and 
description, and of which New Zealand is the type locality. 



BY C. HEDLBY. 303 

Hemidonax dactylus, n.sp. (Plate xxxi., fig. 13.) 

Shell elongate ovate, somewhat the shape of a date stone, solid, a little com- 
pressed, anterior end three-fifths of total length, linguiform, posterior side short 
and broad with rounded angles. Colour pale buff suffused with pale lavender 
towards the beaks; on this ground are irregularly disposed, narrow, radiating, 
brown lines sometimes in broken lengths and sometimes in jagged forks; the 
interior has a white margin round a livid brown centre. Sculpture: ii-regular 
concentric growth lines, waved at the passage of the radials, overi'un the whole 
shell; the centre of the disk is traversed by about eighteen rather coarse, evenly 
spaced, radial riblets, the anterior third and posterior sixth being left smooth; 
as striae these radials are noticeable on the umbonal area; independent of these 
radii is another series of closely packed microscopic hair lines which overrun the 
whole shell except a narrow anterior and posterior segment. On the inner 
ventral margin are twenty-two strong denticules, followed, after a few transitional 
bifid teeth, by a series of sixteen • smaller teeth at the anterior end; all these are 
visible from outside when the valves are clenched. The beaks make a close 
approach to one another. The ligament is wholly external, whereas in H. pictus 
it has made progress in descent into the hinge plate. Length, 27; height, 16; 
breadth of conjoined valves, 12 mm. 

Hah. — N.S. Wales: Kiama (type, R. Helms), Twofold Bay (self), Gerrin- 
gong (Mr. Varney Parkes), Sydney (Miss L. Parkes), Port Stephens (Har- 
graves Coll.). 

This species is readily distinguished from H. pictus which inhabits the 
same zoological subregion, by its slender form and rounded posterior margin. It 
is indeed the narrowest of the genus. 

« Dr. Lamy has given a valuable summary of the genus Hemidonax (Journ. 
de Conch., Ixii., 1917, p. 264) with complete bibliography and fig-ures of the 
hinge of H. cardioides. Though some previous writers have counted two or 
three species Dr. Lamy regarded Hemidonax as a monotypic genus. On the 
contrary I point to four distinct Australian species in addition to the novelty 
described above. 

To begin with, Hemidonax donaciformis should be rejected from the Aus- 
tralian fauna because the original figure (Schroeter, Einleit. Conch., iii., 1786, 
p. 68, ii., PL vii., fig. 14) shows a subtriangular shell, inequilateral, with a very 
short posterior side, grooved across the whole breadth of the valve; this is quite 
unlike any of the shells figured under the name of donaciforme by Wood, Sower- 
by, Reeve, Chenii or Romer. 

The first Australian species was named Donax cardioides by Lamarck (An. 
s. vert, v., 1818, p. 550). It is an equilateral shell, furrowed across the whole 
breadth of the disk and with an orange spot inside; its type locality is the Nuyts 
Archipelago in the Great Australian Bight. Later, it was figxired by Delessert 
(Recueil, 1841, PI. vi., fig. 14). Prom Papuan material it was figured by Quoy 
and Gaimard (Zool. Astrolabe, iii., 1835, PL 81, fig. 17-18) and again by Hanley 
(Cat. Recent Shells, 1843, p. 82, PL 13, fig. 9). Probably this is the species 
which Reeve (Conch. Icon., ii., 1844, PL 5, fig. 25) and Romer (Conch. Cab., 
1869, p. 109, PL 14, fig. 16, 17) misidentified as Cardium donaciforme. But the 
Hemidonax common in South Australia and Victoria reported as cardioides by 
Tate (Trans. Roy. Soe. S.A., ix., 1887, p. 86) and by Pritehard and Gatlifif 
(Proe. Roy. Soc. Victoria, xvi., 1903, p. 119) appears to me to be a second 



304 STUDIES ON AUSTRALIAN MOLLUSOA^ XIV., 

distinct species and has been described by Gatliff and Gabriel as H. chapmam 
(Viet. Nat., xl., 1923, p. 10, PI. ii.). 

The third Australian species is the Cardium australiense of Reeve (Conch. 
Icon., ii., 1844, PI. v., fig. 24). This is also equilateral but is differentiated from 
the previous species by being smooth on the anterior half. Brazier wrote (These 
Proceedings, v., 1880, p. 487) that it is rare in South Australia. Perhaps it has 
not been taken there since Dr. J. B. IJarvey obtained the type at Port Lincoln 
about 1842. It seems to me to be the species figured as Cardium donaciforme by 
Sowerby (Conch. lUustr. Cardium, 1840, p. 6, PL vi., fig. 27). Under the name 
of Donacicardium australiense it was reflgured by Vest (Jahrb. deut. mal. gesell., 
iii., 1876, p. 291, PI. 10, fig. 1). 

A third Australian species was published by Try on (Am. Joum. Conch., vi., 
1870, p. 23, PI. 1, fig. 1) as Donax {Serrula) pictus. The habitat was unknown 
to that writer; specimens before me show pictus to range from Sydney north to 
Port Curtis, Queensland. Like australiense it has a smooth anterior! end but is 
lefes equilateral than that and when full grown develops a posterior heel; it is 
very solid and one of the largest of the genus, reaching a length of 40 mm. 

Dr. W. H. Dall (Trans. Wagner Inst., iii., 1900, p. 963) has named Cardium 
donaciforme Spengler as the type of this genus. But von Martens noted (Zool. 
Eecord, 1870, p. 172) that according to Moreh the type of Hemidonax is Donax 
pictus. 

I repeat my conclusion (These Proceedings, xxx., 1906, p. 540) that Herm- 
donax is nearly related to Gyam^iomactra) this affinity is more apparent in the 
young stages of Hemidonax. 

Dr. Lamy rejects my suggestion that Hemidonax might enter the Crassa- 
telHtidae and restores it to Donacidae. I feel more confident that the entire 
pallia! margin should exclude it from the Donacidae than it should provide ad- 
mission to Crassatellitidae. It is not clear why Dr. Lamy disallowed Fischer's 
reference to Tancrediidae. 

Cardium oxygonum Sowerby. 

Cardium oxygonum, Sowerby, Conch. Illustr. Cardium, part 47, fig. 9, 1833; 
Sowerby, Proc Zool. Soc, 1840, p. 107; Eeeve, Conch. Icon., ii., 1845, PI. 16, 
fig. 77; Try.i>n, Am. Joum. Conch., vii., 1872, p. 262; Shirley, Proc. Roy. Soc. 
Queensland, xxiii., 1911, p. 95. 

Little is recorded of the distribution of this species. The following may be 
noted. — Queensland: Burnett River (Shirley), Caloundra (T. H. Johnston); 
N.S. Wales: Trial Bay (C. Laseron), Middle Harbour (Miss L. Parkes). 

Opisocardium guichardi Bernard!. 

Cardium guichardi, Bemardi, Journ. de Conch., vi., 1857, p. 53, PI. 2, fig. 4. 
Unrecorded for Australia. I collected several separate valves on a beach at 
the mouth of the Annam River, North Queensland. 

PiTARiA INCONSTANS, n.sp. (Plate xxxi., fig. 14, 15, 16.) 

Shell small, solid, much inflated, roundly subtriangular, posterior end twice 
as long as the anterior, variable in the extent to which the posterior end is ab- 
breviated or prolonged. Colour buff, fading to white at the margin and darken- 
ing to pale orange at the umbo; within pale yellow. Sculpture: fine concentric 
hair lines imposed on irregularly spaced larger undulations. Umbo protuberant. 



BY C. HEDLEY. 305- 

apex incurved. Lunule limited by a faint impressed line. Length, 14; height, 
12; depth of single valve, 4 mm. 

Hab. — Cairns, Queensland. Abundant on the beach as dead shells (self). 

Judging from the illustration, this is related to Cytheraea acuminata Sowerby 
(Thes. Conch., ii., 1853, p. 633, PL 136, fig. 178) but differs by being longer in 
proportion to height and by the shorter anterior end. 

Antigona levukensis Smith. 

Venus (CMone) levukensis, Smith, Chall. Report, Zool. xiii., 1887, p. 128, 
PI. 3, fig. 6. 

This species has not been recorded from Australia. In July, 1916, I col- 
lected several dead valves on the beaches of Lizard and of Eagle Islands, 

Bassina pachyphylla Jonas. 

Venus pachyphylla, Jonas, Archiv fur Naturgeschichte, 1839, v., (i.), p. 
344, PL ix., figs. 6, 7. 

This proves to be the earliest name for the species now known as Venus 
paucilamellata Dunkier or Venus alatus Reeve. The illustrations given by Jonas 
are excellent and readily recognisable though his locality ''Patria mare chinense" 
is of course erroneous. At the same pKee the Australasian species. Cassis 
hicarinata and Struthiolaria sulcata, are also ascribed by him to China. 

Except that this name has been indexed in the Thesaurus and in Jay and 
Paetel, it has been completely lost by modern authors. Even the paper in which 
it appeared has escaped the Royal Society Catalogue. 

An unrecorded synonym of the related Bassina {Callanaitis) disjecta Perry, 
seems to be Venus agaricoides (Brown, MS. Cat. Bligh Shells, 1822, p. 45, No. 
764 and Mawe, Linn. Syst. Conchology, 1823, p. 46). This comparison also oc- 
curred to Tenison Woods who, in his Census, remarked that the undulose friUs 
of Venus lam.eUata, striate on the lower side and of a pink colour, reminded him 
of the gills of a mushroom. 

While examining Lamarck's collection in the Geneva Museum, I noticed 
that his Venus lamellata, var. 2 (Lamarck, p. 598) was not that species but the 
ISTew Zealand B. yatei Gray. 

P'seudaroopagia botanica Hedley. (Plate xxxi., figs. 17, 18, 19.) 

Tellina fimbriata, Romer, Monogr. Tellina, 1872, p. 84, PL 23, figs. 4, 5, 6; 
not T. fimbriata, Hanley, 1844. — Tellina decussata, Angas, Proe. Zool. Soc, 1877, 
p. 191, not T. decussata of Wood or of Lamarck. — Pseudarcopagia botanica, 
Hedley, Journ. Roy. Soc. N.S.W., li., 1918, p. M. 25; May, Proc. Roy. Soc. 
Tasm., 1919, p. 68 and Check-list Moll. Tasmania, 1921, p. 25, and Illustr. Index 
Tas. Shells, 1923, PL xi., fig. 7. 

The specimen here figured was taken by myself in Middle Harbour, the 
colour is dull white passing gradually into pale orange at the umbo, the length 
is 29 and the height 27 mm. 

On the mainland I trace P. botanica from Caloundra, Queensland (Kesteven) 
to Twofold Bay, N.S. Wales (self). In Tasmania Mr. W. L. May finds^that it 
reaches the east and south coasts and is particularly abundant in the Purneaux: 
group. T. victoriae and T. botanica appear as geminate species {see discussion, 
These Proceedings, xxxv., 1910, p. 416 and D. S. Jordan, Bijd. Dierk., xxii., p, 
175-178), representing one another on opposite coasts. 



306 STUDIES 01^ AUSTRALIAN MOLLUSCA^ xiv., 

Pseudareopagia was instituted by Bertin (Nouv. Arch. Mus., (2), i., 1878, 
p. 264) for a small group of southern Tellinids easily defined by having the 
sinus confluent with the pallial line. He placed there T. decussata Lamarck from 
Australia and T. disculus Deshayes from New Zealand. He expected that the 
group would soon be increased by the addition of other species, but authors 
generally have ignored the matter and the jDresent is the first extension of 
Pseudareopagia. 

The first species, which I now nominate as type of the genus, was described 
as Tellina deciissata by Lamarck (An. s. vert., v., 1818, p. 532) the type locality 
being King Greorge Sound, W.A. As Hanley suggested this type is probably 
illustrated by the unnamed fig. 2, PL 292 of the Eneyclopedie methodique. Three 
years before Lamarck wrote, a very different shell, possibly a Semele, had been 
described by Wood (General Conchology, i., 1815, p. 190, PI. 43, fig. 2, 3 and 
Index test., 1825, p. 22, PI. 4, fig. 81) as Tellina decussata. On this account, 
Gatliff: and Gabriel (Vict. Nat., xxxi., 1914, p. 83) renamed the Australian shell 
Tellina victoriae. 

Peteicx>la pseudolima Soviverbie. 

Petricola pseudolima, Souverbie, Journ. de Conch., x., July, 1862, p. 231, PL 
9, fig. 1; Venerupis texta, Sowerby, Conch. Icon., xix., November, 1874, PI. 2, 
fig. 9. 

The type of this species was obtained from He Art in New Caledonia by 
Pere Montrouzier. V. texta, which has not hitherto been recognised as conspecifie, 
was reported from North-East Australia. I have taken it at Green Island, near 
Cairns. 

PSAMMOBIA BOSSITERI CrOSSe. 

Psammohia rossiteri, Crosse, Journ. de Conch., xxi., 1873, pp. 66 and 129, 
PI. v.. fig. 6. 

The range of this New Caledonian species is now extended to Australia by 
a single valve which I picked up on the beach of Lizard Island, Queensland, 
19/7/16. It was determined by comparison with the type of Crosse which, with 
Crosse's autograph label, is now in the Australian MvTseum. 

Argonauta bottgeri Maltzan. (Plate xxxi., fig. 11.) 

Argonauta hottgeri, Maltzan, Journ. de Conch., xxix., 1881, p. 163, PI. 9, 
fig. 7; Smith, Ann. Mag. Nat. Hist., (5), xxi., 1887, p. 409, PL 17, fig. 1-6; 
Brazier, Cat. Austr. Cephalopoda, 1892, p. 3; Hoyle, Proc. Roy. Phys. Soc. 
Edinburgh, xiii., 1897, p. 365; Hidalgo, Rev. R. Acad. Cien. Madrid, 1905, p. 
9; Hall, BulL Mus. Comp. ZooL, xliii., 1908, p. 226; Berry, Bull. Bureau 
Fisheries, xxxii., 1914, p. 277, text-figs. 3-7, PL 48, fig. 5; Massey, Cephal. Terra 
Nova Exped., 1916, p. 143; fig. 1, 2. 

Though reported from Australia, no definite locality on our coast has been 
indicated for this species. It is, therefore, interesting to record that an empty 
shell, maj. diam. 38, min. diam. 21, breadth 16 mm., was collected 18/5/19 on 
Maroubra beach near Sydney, after a heavy gale, by Mr. F. A. McNeill. Perhaps 
the record by Blainville (Diet. Sei. Nat., xliii., 1826, p. 213) of Argonauta 
crassieosta from New Holland was based on the present species. Our shell has 
a general resemblance to A. oiuenii, but, as Smith has shown, the rough granular 
surface provides a good recognition mark to distinguish A. hottgeri from the 
rest of the genus. 



BY C. HEDLEY, 307 

The "Albatross" took it off the Hawaiian Islands and the "Terra Nova" met 
it off the New Zealand coast in about the latitude of Sydney. This distribution 
suggests that it is a tropical species which has drifted to our beach. 

Emarginula bajula Hedley. 

Emarginula bajula, Hedley, These Proceedings, xxxviii., 1913, 276. 

This Sydney species has been reduced by Dr. Thiele (Conch. Cab., 1915, p. 
86) to a synonjon of the Western Australian E. dilecta A. Adams. It was only 
distinguished from E. dilecta after my examination of the type of E. dilecta in 
the British Museum. 

With all deference to the valued opinion of the Berlin conehologist, I submit 
that his conclusion is based on a misapprehension of thq identity of E. dilecta. 
By an interrogation mark, Dr. Thiele expresses doubt whether that species is re- 
presented by a figure of the type given by the author. Dr. Thiele gives an 
original figaire (Plate 10, figs. 13, 14), different in shape, longer in proportion to 
breadth and with twice as many radials as the type. It is evident that a species 
different from E. dilecta and probably E. bajula is here shown. 

Perhai3s E. bajula should be transferred to Emarginella. 

Emarginula devota Thiele. 

Emarginula devota, Thiele, Conch. Cab., ii., abth. 4a, 1915, p. 81, PL 9, figs. 
27, 28. Type locality. Port Jackson. 

This species is an addition to our fauna. 

Emarginula hedleti Thiele. 

Emarginula hedleyi, Thiele, Conch. Cab., ii., abth. 4a, 1915, p. 74, PI. 9, 
figs. 8, 9. Type locality. Port Jackson. — E. dilecta, Angas (not Adams), Proc. 
Zool. Soc, 1867, p. 219. — E. Candida, Angas, Proc. Zool. Soc, 1877, p. 189; not 
E. Candida A. Adams, Thes. Conch., iii., 1863, PI. 246, figs. 45, 46; not E. 
Candida Thiele, op. cit., PI. 8, figs. 15, 16, 16a. 

This Peronian shell has hitherto been confused with its geminate E. Candida 
A. Adams from Port Adelaide. Prom that it can be distinguished by shorter 
length, higher arch of back, and blunt, submedian apex. 

EURYTROCHUS DANIELI CrOSSe. 

Clanculus danieli, Crosse, Journ. de Conch., x., 1862, p. 407, PI. 13, fig. 5. — 
Trochus danieli, Pischer, Journ. de Conch., xxiii., 1875, p. 49 and Coq. Viv., 1878, 
p. 326, 417, PL 102, fig. l.—Gibhula danieli, Pilsbry, Man. Conch., xi., 1889, p. 
229, PL 32, figs. 57, 58, 5Sa..—aihhula filosa, Garrett, Proc. Cal. Acad. ScL, iv., 
1872, p. 201. 

Unrecorded for Australia. I found a single shell on the beach of Lizard 
Island, 18/7/16. 

EurytrocTius was proposed by Fischer (Coq. Viv., Troque, 1879, p. 417) for 
Trochus coxi Angas, danieli Crosse, lehmanni Menke and reevei Montrouzier. He 
afterwards selected (Man. de Conch., 1885, p. 824) E. danieli to serve as type. 
I think that T. lehmanmi and hicarinata Adams (= coxi Angas) do not agTee 
with this type and suggest that they be withdrawn from EurytrocTius and trans- 
ferred to Minolia. In their place I add the following Australian species to 
Eurytrochus, — Gibbula eoncinna Dunker, GHbbula maecullocJii Hedley, Trochus 
plehejus Philippi and Gibbula strangei A. Adams. 



308 STUDIES ON AUSTRALIAN MOLLUSCA^ xiv., 

Calliotbochus striatulus Garrett. 

Trochus striatula, Garrett, Proe. Calif. Acad., i., 1857, p, 102. — Monilea^ 
striatula, Paetel, Cat., i., 1887, p. 568. — Margarita striatula, Pilsbry, Man. o£ 
Conch., xi., 1889, p. 249, PL 61, figs. 19, 20; Melvill and Standen, Journ. de 
Conch,, viii., 1895, p. 126. — Turbo pliasianella, Deshayes, Conehyl. de L'ile de 
JReunion, 1863, p. 74, PL 9, fig. 13-14. — Trochus pliasianella, Fischer, Journ. de 
Conch., xxii., 1874, p. 136 and xxiii., p. 49; Coq. Vivant, 1879, p. 363, PL iii., 
fig. 4 and p. 418. — Gihhula phasianella, Hedley, Mem. Austr, Mus., iii., 1899, p. 
405; not Turbo pJiasianella C. B. Adams, Panama Shells, 1852, p. 195. 

Unrecorded for Australia. I found a few specimens on the beaches at 
Eagle Island, at Two-Isles and Rocky-Isle oif Cape Flattery, Queensland. I 
have also seen it from Lord Howe Island. 

MONILEA BELCHERI Philippi. 

Trochus belcheri, Philippi, Zeitsch. Malak., vi., 1849, p. 148; Conch. Cab., 
1852, p. 302, PL 44, fig. 3; Watson, ChalL Rep. ZooL, xv., 1886, p. 71. 

This s]')ecies has not been recorded from Australia. I found a few speci- 
mens in July, 1916, on Rocky-Isle off Cape Flattery, Eagle Island and Lizard 
Island. 

Astraea sirius Gould. (Plate xxx., fig. 1.) 

Turbo (Stella) sirius, Gould, Proc. Bost. Soc. Nat. Hist., iii., 1849, p. 83; 
Hedley, These Proceedings, xxxviii., 1913, p. 283. 

The common Sydney Astraea was identified by Angas (Proc. ZooL Soc, 
1867, p. 214) as ZJvanilla tentoriiformis Jonas. On re-examination, and with a 
distrust of the identification of Angas, bred of experience, I now consider that 
the species of Jonas is a form of South- West Australia, differing by the features 
of the base from ours. In this connection it is noteworthy that most of the 
marine Australian gasteropods handled by Jonas, such as Astraea aurea, Fusinus 
dunheri, F. philippi and Cassis bicarinata, were of South- West origin. 

In this case our shell will resume the name given to it by Gould. 

Following are notes on the animal, to accompany my drawing of the living 
mollusc. Where the water is clear the species is abundant on rocky ground at 
half-tide level, near Sydney. The epipodial fringe is here reduced to a ragged 
ribbon extending from behind the ocular tentacle to a point half-way to the 
end of the operculum. There are on each side four cirrhi, three round the 
operculum and one anterior to it; these are short stumpy little tags. Frontal 
lobes quite absent. It is suggested that Astraea has degenerated from the Turbo 
type in the direction of Nerita. The elongate paucispiral operculum is further 
from the primitive multispiral type than that of Turbo. The loss of the frontal 
iobe also indicates a recession from the primitive epipodial development. 

Kesteven (These Proceedings, xxvii., 1902, pp. 3-5, figs. 2, 4, 5) has described 
and figured the operculum and radula. 

Nerita melanotragus Smith. (Plate xxx., figs. 4, 5.) 

For previous notes see These Proceedings, xli., p. 706. The ova of this 
species has not been described previously. They are deposited on shells or 
stones in small irregular groups of ootbeca. Each nidus is from one to two 
millimetres long, appearing as a small, oblong, hard, white scale. On magni- 



BY G. HEDLEY. 309 

fieation, the cluster is resolved into sevei'al hundred, minute, spherical eggs 
closely packed in a connecting membrane. The whole assumes the shape of an 
invea-ted boat attached by the margin. The eggs project from the upper sur- 
face; the under side is smooth and glossy. I collected the example figured in 
Middle Harbour, 9th August, 1917, under stones at mean low tide. 

Patella onychitis Menke. 

Patella onychitis, Menke, Moll. Nov. Holland Spm., 1843, p. 34. 

This name has not been recognised by modern students. The suggestion is 
now put forward that it may be a synonym of Patelloida septiformis Quoy and 
Gaimard, Zool. Astrolabe, 1834, PL 71, fig. 43. 

Cheilea undulata Bolten. 

Patella undulata, Bolten, Mus. Bolt., 1798, p. 4 for Martini, Conch. Cab., 
i., 1767, PI. xiii., fig. 119, 120. — Mitrularia neptuni, Schumacher, Essai Nouv. 
Syst., 1817, p. 183 for the same figure of Martini and Rumphius, Mus. Tab. 40, 
fig. P.Q. — Calyptraea dormitoria Reeve, Conch. Icon., xi., 1858, PI. ii., fig. 5a, 5b. 

This seems to be the proper name for a shell erroneously recorded from 
Australia by various authors as Calyptraea equestris. Both Lamarck and Linne 
confused two species in their references to Patella equestris. According to 
Hanley (Ips. Linn. Conch., 1855, p. 414) that species is characteristically shown 
by Gualteri, PI. 9, fig. x. 

I have seen C. v/ndulata from the Sow and Pigs reef in Sydney Harbour 
and have gathei-ed it at intervals between the Capricorn Islands and Torres 
Straits, 

Crepidula aculeata Gmelin. (Plate xxx., fig. 6.) 

Patella aculeata, Gmelin, Syst. Nat., xiii., 1791, p. 3693. 

This species is one of the few that range round the whole world. In Sydney 
Harbour it lives on the under side of stones in the Eklonia zone, at or near low- 
tide level. The animal is uniform buff, the head and mantle with numerous 
opaque white spots. The hind part of the foot is fast to the shell, but the 
fore foot is extremely mobile; it can reach out far enough to right itself when 
the animal is turned over on to its back; sometimes the anterior margin appears 
truncate, at other times, and as in my drawing, it is emarginate. The muzzle 
has two pxojecting lobes which are constant in all attitudes. The tentacles are 
moderately long, with eyes on the outer side of their swollen base. Intromittent 
organ behind the right tentacle and usually doubled back. Body much com- 
pressed, the sides extending in thin flanges. At the anterior end the mantle 
forms a short siphon. Ctenidium extremely large, roofing the whole chamber 
behind the head. 

Cerithium dorsuosum Menke. (Plate xxxi., fig. 12.) 

Cerithium tuberculatum, Monke {not Lamarck), Moll. Nov. HoU. Spm., 
1843, p. 19. — Cerithium dorsuosum, Menke, Zeitschr. Malak., i., 1844, p. 60; not 
Cerithium dorsuosum, Sowerby, Thes. Conch., ii., 1855, p. 867, PL 181, figs. 138, 
139, 140. — Cerithium tuberculatum, Sowerby (not Linne), Thes. etc., p. 870, PL 
182, figs. 164, 165, not 162. — Cerithium jannellii, Hombron and Jacquinot, Voy. 
Pole Sud, 1842-1853, Moll., PL 24, figs. 19, 20; Rousseau, MoU., 1854, p. 104.— 
Cerithium variegatum var. janellei, von Martens, Zool. Ergeb. Reis. Niederl. 



310 STUDIES ON AUSTRALIAN MOLLUSOA^ xiv., 

Ostind., Bd. iv., 1897, p. 171. — Cerithium janellei vax. moniliferwm, Shirley, 
Proc. Roy. Soc. Queensland, xxix., 1917, p. 110. 

This species is generally distributed on the coast of Queensland as the 
following records show: St. Crispin's Reef on the Outer Bariier (McCulloch), 
Eagle, Hope and Palm Islands (Hedley), Facing Island (Kestfven), Caloundra 
(Johnston). 

Cerithium vignali Sowerby. 

Cerithium vignali, Sowerby, Ann. Mag. Nat. Hist., (8), ix., February, 1912, 
p. 237, fig. 1; not Terehralia vignali, Cossmann, Ess. Paleoconeh. comp., vii., 
1906, p. 242, PL X., fig. 14. 

New to Australia. One specimen collected by myself dead on the beach at 
Lizard Island, North Queensland, 19/7/16. 

Ataxocserithium abbreviatum Brazier. 

Cerithium abbreviatum. Brazier, These Proceedings, i., 1877, p. 316; Melvill 
and Standen, Journ. of Conch., viii., 1897, p. 409. — At axo cerithium abbreviatum, 
Hedley, Ree. Austr. Mus., iv., 1901, p. 126, PI. xvii., fig. 24; Hedley, These 
Proceedings, xxxii., 1907, p. 481; Not Cerithium abbreviatum Deshayes, f,de 
Cossm.scnn. —AtaxoceritMum brazieri, Cossmann, nom. mut., Essai Paleoconeh. 
Comp., vii., 1906, p. 92, footnote. 

This species has been reported from Katow, Papua, from Lifu, Loyalty 
Islands and from Masthead Island, Queensland. Noticing that Brazier's name 
was preoccupied for an Oligocene fossil, Cossmann has changed it as above. He 
does not, however, give the original reference on which Deshayes' name depends 
and I am unable to recover it. Apparently Cossmann's synonymy is based on 
Melania abbreviata Defrance (Diet. Sci. Nat., xxix., 1823, p. 467) which is in- 
effective. Cossmann also (p. 91) extends this genus to New Zealand in the 
person of a fossil, Ataxocerithium huttoni {Cerith. cancellatu^m, Hutton, not. 
Lamk. ) . 

BiTTiUM lacertinum Gould. 

Cerithium lacertinum, Gould, Proc. Bost. Soc. Nat. Hist., vii., 1861, p. 386; 
Gould, Otia Conch., 1862, p. 141; Sowerby, Conch. Icon., xv., 1865, PI. xviii., 
fig. 129; Tryon, Am. Journ. Conch., ii., 1866, p. 286; Ten. Woods, Proc. Linn. 
See. N.S.W., ii., 1878, p. 258; Kobelt, Conch. Cab., 1898, p. 248, PL 43, fig. 
7. — Bittium granarium, Angas, Proc. Zool. Soc, 1867, p. 208; Watson, Chall. 
Rep., ZooL XV., 1886, p. 539; Hedley, Proc. Linn. Soc. N.S.W., xxxix., 1915, p. 
718, PL 77, fig. 6 (from Gould's type) ; Not Cerithium granarium Kiener, Coq. 
Viv., 1842, p. 72, PL 19, fig. 3. 

Misled by Angas, who asserted that lacertinum was the same as granarium 
from Western Australia, subsequent writers have reduced Gould's name to 
synonymy. Having reconsidered the matter, I find that lacertinum consistently 
differs from granarium by less height, greater proportionate breadth and more 
closely packed grains. I conclude that Gould's lacertinum should be resumed for 
the Sydney shell and that B. granarium doas not reach the Pacific coast. The 
species recorded as B. granarium from New Zealand is, in my opinion, wrongly 
identified. 

Triphora cancellata Hinds. 

Triphoris cancellatus, Hinds, Ann. Mag. Nat. Hist., xi., 1843, p. 18; Zool. 
Sulphur, 1844, p. 28. PL 8, fig. 6. 



BY G. HEDLET. 311 

New to Australia. Collected by self, 3/8/16, dead on the beach at Rocky- 
Isle off Cape Flattery, North Queensland. 

Triphora concinna Hinds. 

Triphoris concinnus, Hinds, Ann. Mag. Nat. Hist., xi., 1843, p. 20; Zool. 
Sulphur, 1844, p. 31, PI. 8, fig. 20. 

New to Australia. Collected by self, 5/8/16, dead on the beach at Two- 
Isles, off Cape Flattery, North Queensland. 

Triphora ducosensis Jousseaume. 

Mastonia ducosensis, Jousseaume, Ball. Soc. Malac. France, i., 1884, p. 251, 
PI. 4, fig. 10. 

Unrecorded for Australia. I took a few specimens on the beach at Two- 
Isles off Cape Flattery. 

Triphora monilifera Hinds. 

Triphoris monilifer, Hinds, Ann. Mag. Nat. Hist., xi., 1843, p. 19; Zool. 
Sulphur, 1844, p. 30, PI. 8, fig. 14. 

New to Australia. Taken by self, 5/8/16, dead on the beach at Two-Isles, 
off Cape Flattery, North Queensland. 

Tukritella joannae^, nom. mut. 

Turritella (Colospira) crenulata, Donald, Proe. Malac. Soc, iv., 1900, p. 52, 
PL 5, fig. 2, 2b; not Turritella cr&nulata Nyst, Descrip. Coquilles Polyp, foss. 
Belgique, 1843, p. 399, PI. 37, fig. 6. — Turritella reevei, Cossmann, Essais Paleo- 
coneh. compar., ix., 1912, p. 119; not Turritella reevei, Dautzenberg and Fischer, 
Journ. de Conch., liv., 1907, p. 163. 

It was noted by Cossmann that a name given to a Sydney shell had served 
for more than half a century to distinguish an Oligocene fossil from Belgium. 
Unfortunately he mistook the author of the recent shell who was not, as he sup- 
posed, Mr. Lovell Augustus Reeve, but who was Mrs. G. B. Longstaff, formerly 
Miss Jane Donald. Also he overlooked the fact that the name of Turritella 
reevei was already in use for a shell from Annam. This last error enables me 
to carry out Cossmann's intention to name it after the talented lady who des- 
cribed the Australian shell. 

Cymatium pyrum Linne. 

Much learning and skill were brought by Sylvanus Hanley to his analysis 
of the names in Linnean conehology. Yet the principles which guided him were 
so different from the rules which now prevail that some of his conclusions may 
be reviewed with advantage. 

Murex pyrum was founded by Linne (Syst. Nat., x., 1758, p. 749) on the 
following fignires. — "Rumf . Mus. t. 26, f . E . f The "gedroog*te peer" was de- 
termined by von Martens (Gedenkboek, 1902, p. 116) to be T. pyrum; "Gault. 
test. t. 37, f. 7." The T. pyrum of authors, f,de Hanley; "Argenv. conch, t. 13, 
f. 7." T. sarcostoma of Reeve, fide Hanley; "Regenf. t. 6, f. 60." T. pyrum, 
fide Deshayes (An. s. vert, ix., 1843, p. 633) ; "Regenf. t. 5, f. 50." T. clavator, 
fide Hanley. 

The proposal of Hanley (Ips. Linn. Conch., 1855, p. 290) was to discard the 
original account of the tenth edition and to decide the identity of M. pyrum by 
the twelfth edition where different characters point to a different shell. On this 
ground he determined the Linnean M. pyrum as being that shell which Reeve 
figured as Triton clavator. 



312 STUDIES 02sr AUSTRALIAN" MOLLUSCA^ xiv., 

On tlie contrary I now suggest that, since Linne included three different 
species under Murex pyrum, the name should be indicated by the majority of 
citations and that early writers, such as Dillwyn, were justified in eliminating 
the other two from the residual which they regarded as Murex pyrum. The rul- 
ing of Hanley has been generally disregarded. 

A fine example of Cymatium pyrum, 115 mm. in length, was taken by Dr. 
H. L. Kesteven on the beach of Facing Island, Port Curtis, Queensland. This 
is the first appearance of the species in Australia, for in recording G. pyrum 
(These Proceedings, xxxiv., 1909, p. 451) I had followed Hanley's guidance. An 
unquoted fig-ure of C. pyrum is Hirase, Conchological Magazine, June, 1907, PL 
xii., fig. 87. 

Both the related species, G. sarcostoma Reeve and G. clavator Dillwyn (t=! 
formosus Perry), occur in Queensland. 

Natica bougei Sowerby. 

Natica hougei, Sowerby, P'roe. Malac. Soc, viii., 1908, p. 17, PL 1, fig. 3. 
Some specimens I found east up' on the beach at Rocky-Isle off Cape Flattery, 
3/8/16, enable me to add this species to the io^tralian list. 

POLINICES PETIVERIAlSrA Recluz. 

Natica petiveriana, Recluz, in Chenu, Illustr. Conch. Livr., 12-13, 1843, 
Natica, PL 2, figs. 5-9. — Natica hicolor, Philippi, Zeitschr. Malak, 1848, p. 156 
and Conch. Cab,, p. 43, PL 6, fig. 4; Reeve, Conch. Icon., ix., 1855, PL iv., 
fig. 17. 

This earlier name of Recluz seems to have been overlooked. The species 
occurs in Queensland and in North- West Australia. 

POLiNiOES SOEDIDUS Swainson. 

Natica sordida, Swainson, Zool. Illustr., 1st ser., vol. ii., PL 79, central 
fignires (Nov. ?), 1821. — Natica plumhea, Lamarck, Anim. s. vert., vi., (2), p. 
198, April, 1822; Delessert, ReeueU, 1841, pi. 32, fig. 10. 

It has long been recognised that sordida and plumhea were names for the 
same shell, but which has priority of publication has been left undecided. In 
1844 Philippi wrote of sordida (Enum. mollusc. Siciliae, ii., p. 140), "Quod 
nomen prius datum est ? Swainsonii an Lamarckii 1" but without reaching a 
decision. 

Recently it was ascertained by Mathews (Birds of Australia, vii., part v., 
1919, p. 468) that the "First Series" of the "Zoological Illustrations" were issued 
in monthly parts with six plates to a part and that part four was dated "Jan. 
1, 1821." From this data it can be reckoned that Natica sordida was published 
towards the end of 1821 and has definite priority over plumhea. 

Epitonium bavayi de Boury. 

Scala (Girsotrema) bavayi, de Boury, Joum. de Conch., Ix., 1913, p. 280, 
PI. 10, fig. 6, previously considered to be Scala multiperforata. 

New to Australia. Collected by self, dead on the beaches of Rocky-Isle and 
Two-Isles off Cape Flattei-y, July, August, 1916. 



BY C. HEDLEY. 313 

Cypraea cumingii Sowerby. 

Cypraea cumingii, Sowerby, Conch. lUustr. Cypraeidae, 1832, p. 8, PI. 1, 
fig. 5; Hidalgo, Monogr. Cypraea, 1906, p. 325. 

This species has not hitherto been observed in Australia. One specimen was 
collected in 1918 by Mr- 4. R. McCulloeh in 20 fath. off the Endeavour Reef, 
Queensland. 

Terebra funiculata Hinds. 

Terebra funieulata, Hinds, Proc. Zool. Soc, 1843, p. 153; Zool. Sulphur, 
1844, p. 33; Thes. Conch., i., 1844, p. 168, PL 43, fig. 63; Tryon, Man. Conch., 
vii., 1885, p. 29, PI. 9, fig. 60 (PL 12, fig. 48 is T. archimedis, Deshayes, Proc. 
Zool. Soc, 1859, p. 314, fide Reeve, Conch. Icon., xii., 1860). 

Unrecorded for Australia. Three specimens collected dead on the beach at 
Lizard Island, North Queensland by self, 19/7/16, and identified from a New 
Caledonian specimen named by Mr. Fulton. 

Engina mundula Melvill and Standen. 

Engitia mundula, Melvill and Standen, Journ. of Conch., viii., Oct., 1895, p. 
105, PL 2, fig. 6. 

Unrecorded for Australia. I collected several specimens alive under stones 
on the Outer Barrier Reef near Crescent Reef, 30/7/16. 

MiTRA FULVOSULCATA Melvill. 

Mitra fulvosulcata, Melvill, Journ. of Conch., v., 1888, p. 287; Melvill and 
Standen, Journ. of Conch., viii., Oct., 1895, p. 101, PL 3, fig. 32. 

Unrecorded for Australia, one specimen was collected on the Outer Barrier 
Reef, off Cairns, by Douglas Pitt. 

Pyrene versicoi/Or var. atladona Duclos. (Plate xxx., figs. 2, 3.) 

Columhella atladona, Duclos, Hist. Nat. Coq. Univ., 1840, PL i., figs. 11-12. 

The type of Columhella versicolor Sowerby, came from the island of Anaa 
in the Paumotus; a serviceable name for the Sydney expression of this tropical 
species is var. atladona Duclos. A form from Western Australia, usually united 
with versicolor, but sufficiently distinct for specific recognition, is Pyrene arane- 
osa Kiener, 1841 {■= script a Lamarck, 1822 {not Linne, 1758) = hidentata 
Menke, 1843). To contain this group Dr. Dall has proposed (Bull. Mus. Comp. 
Zool., xviii., 1889, p. 187) the subgenus Euplica. 

In Sydney Harbour, this shell may be found on rocks, but prefers the 
fronds of the brown sea-weeds, Sargassum and Eklonia. The dark colour of the 
shell conceals it when nestling among the weed. The foot is very narrow and 
the operculum ungulate. 

Zapra hervieri Pace. 

Columhella peasei, Hervier, Journ. de Conch., xlvii., 1899, p. 368, PL 14, 
fig. 9; not C. peasei von Martens. — Columhella hervieri, Pace, nom. mut., Journ. 
de Coneh., L, 1902, p. 420. 

New to Australia. I found one specimen on the beach at Two-Isles, near 
Cape Flattery, 5/8/16. 



314 STUDIES ON AUSTRALIAN MOLLUSCA^ xiv., 

Zafra obesula Hervier. 

Columbella ohesula, Hervier, Joum. de Conch,, xlvii., 1899 p. 376, PL 14, 
fig. 6. 

New to Australia. I found a dozen specimens in shell drift on the beach 
of Two-Isles, near Cape Mattery, North Queensland, 5/8/16. 

Zafra stepheni Melvill and Standen. 

Columhella pacei, Melvill and Standen, Joum. of Conch., viii., 1896, p. 275, 
PI. 9, fig. 5; not C. paoei E. A. Smith. — Columbella stepheni, Melvill and Stan- 
den, op. cit., p. 407; Heivier, Journ. de Conch., xlvii., 1899, p. 386. 

Unrecorded for Australia. One specimen from the Great Ban-ier Reef off 
Cairns, D. Pitt. 

Drupa tuberculata de Blainville. 

Purpura tuherculata, Blainville, Nouv. Ann. Museum, i., 1832, p. 186, PI. 9, 
fig. 3; Kiener, Coq. Yiv., Pourpre, 1836, p. 22, PI. 5, fig. 10. — Purpura granu- 
lata, Duclos, Ann. Sci. Nat., Mai, 1832, xxvii., p. Ill, PL 2, fig. 9. 

No more definite locality than the original New Holland of Duclos seems 
to have been registered. I gathered it alive on rocks at Eagle Island, 23/7/16, 
and on the Outer Barrier Reef, 30/7/16. 

Tethys angasi Sowerby. (Plate xxxiii., fig. 21.) 

Syphonota keraudreni, Angas, Proc. Zool. Soc, 1867, p. 228; Clessin, Conch. 
Cab., 1899, p. 51, PL 12, fig. 4, 5. — Aplysia keraudreni, Sowerby, Conch. Icon., 
xvii., 1869, PL i., figs. 2a, 2b ; (not Aplysia keraudreni Rang, Hist. Nat. des 
Aplysiens, 1828, p. 69, PL 13). — Aplysia angasi, Sowerby, Conch. Icon., xvii.,. 
1869, PL viii., figs. 35a, 35b; Angas, Proc. Zool. Soc, 1877, p. 190.— Tethys 
angasi, Pilsbry, Man. Conch., xvi., 1896, p. 101, PL 57, figs. 18, 19. 

This, the largest and most handsome of our Sea Hares, was first noted in 
literature by G. F. Angas, who recorded it under the wrong name of Syphonota 
keraudreni; in his revised list of 1877, he substituted for that the name of 
Aplysia angasi Sowerby, type locality. Sow and Pigs Reef, Port Jackson, col- 
lected by J. Brazier. 

By the kind permission of the Council, I can now, for the first time, illus- 
trate this animal, from a beautiful drawing made more than eighty years ago 
by Dr. .James Stuart, then medical officer at the North Head Quarantine Station. 
The series, of which this picture is one, was described by Mr. J. J. Fletcher 
(These Proceedings, xlv., 1921, pp. 609-611). 

The original was doubtless taken at the North Head, this species apparently 
preferring the clear water of the ocean to the estuaries. Our drawing is signed 
J(ames) S(tuart), dated (18)41, and labelled "Aplysia." The animal is evi- 
dently represented life-size, being 280 mm. in total length and 140 mm. in 
height; the slug is shown in full marching order, with the lobes of the mantle 
held aloft, the tentacles exserted and the foot extended. On a dark olive 
glaucous ground a few large black ocelli are scattered; an impressed network 
segregates some large tubercles on the upper surface of the foot; the mantle 
is smooth. The siphon is shown as a funnel the expanded lip of which reaches 
over the parapodial lobes. 



BY C. HEDLEY. 315 

Umbraculum botanicum^ n.sp. (Plate xxxii., fig. 20.) 

Operculatum indicum, Angas, Proe. Zool. Soe., 1867, p, 228; Schmeltz, Cat. 
Mus. Godeffroy, v., 1874, p. 162. — Operculatum aurantium, Whitelegge, Proc. 
Koy. Soe. N.S. Wales, xxiii., 1889, p. 276. — Umbrella indica, von Martens, 
I'orsch. Gazelle, iii., 1889, p. 263. — Umbraculum umbella, Suter, Man. N.Z. Moll., 
1913, p. 549, PI. 23, fig. 14; Iredale, Proc. Malae. Soe, ix., 1910, p. 69; Oliver, 
Trans N.Z. Inst., 1914 (1915), p. 544; Hedley, Proe. Roy. Soe. N.S. Wales, li., 
1918, p. 108. 

Shell shield-shaped, rather flat, thin, covered with a caducous epidermis, 
concentrically striated, irregailarly and faintly radially plicate, margin sinuous. 
Apex on the left side of the centre, prodissoconch erect involute. Colour white, 
interior sear burnt-sienna to lemon yellow. Foot oval, inflated nearly twice the 
length of the shell, exposed surface closely covered with tough, erect, eom|>ound 
tubercles. Sole of foot buff-yellow, tubercles lavender-grey, their interstices 
brown, tentacles and gills buff. Angas reports that a shell collected by himself 
was four inches long. This is an unusual size. 

Hab. — N.S. Wales: Port Jackson (Angas), Maroubra Bay (type) and Byron 
Bay (Hedley); Queensland: Moreton Bay (von Martens), Bowen (Schmeltz); 
Lord Howe Island (Thomson) ; Norfolk Island (Oliver) ; Kermadec Islands (Ire- 
dale) ; New Zealand (Suter). 

This umbrella-shell from the South-west Pacific was determined first by 
Angas as 0. indicum. But it appears sep'arable from U. indicum^, as figured by 
Eydoux Souleyet and Gould, by the larger compound tubercles and from U. 
umbella Martyn or sinica Gmelin by the flatter shell. As the Chinese or Indian 
umbrella-sheUs were not well defined by the eighteenth century eonchologists who 
named them, the best way to avoid confusion seems to be to regard the New 
South Wales form as an independent species until the Oriental species are re- 
examined. 

Cheomodoris BENisrETTi Angas. (Plate xxx., figs. 7, 8, 9, 10.) 

Goniodoris bennetti, Angas, Journ. de Conch., xii., 1864, p. 53, PI. 4, fig. 10. 

This beautiful species is probably the commonest nudibranch in Sydney Har- 
bour. It is sky-blue with a deep orange border to the mantle and spotted with 
scattered blood-red dots. It lurks on the under surface of large stones in the 
Eklonia zone. Transferred to an aquarium it can swim with the sole of the foot 
extended on the surface of the water. The rhinophores are pink. The bran- 
chiae are rose on the inner and outer sides, the rest white; they are furnished 
with nine bipeetinate branches and are completely retractile. 

One of these nudibranchs which I kept in a basin of water laid a coil of 
eggs on November 24th. The ribbon was in a spiral which rapidly enlarged 
and almost formed a closed circle. The orange eggs were imbedded in a trans- 
lucent gelatinous frame twice as high as broad. 

Dr. Bergh has given some anatomical details of this. (Verhandl. k.k. Zool. 
bot. Ges. Wien., 1893, p. 415, PI. iv., f. 12-17). 

Thecidellina maxilla Hedley. (Text-fig. 1.) 

Thecidea maxilla, Hedley, Mem. Austr. Museum, iii., 1899, pp. 508-510, 
text-fig. 57; Mawson, Proc. Linn. Soe. N.S. Wales, xxx., 1905, p. 477.— 
Thecidellina maxilla^ Thomson, Geol. Mag., Dec. vi.. Vol. ii., Oct., 1915, p. 463. 

A single top valve, dredged by myself in 5-8 fathoms, off Mijrray Island, 
Torres Straits, adds another genus to the Australian brachiopoda. 



316 



STUDIES ON AUSTRALIAN MOLLUSCA, XIV. 





Text-fig. 1. Thecidellina maxilla Hedley. 



Fig. 


1 


Fig. 


2. 


Fig. 


3 


Fig. 


4 


Fig. 


5. 


Fig. 


6 


Fig. 


7 


¥ig. 


8 


Fig. 


9 


Fig. 


10 


Fig. 


11 


Fig. 


12 


Fig. 


13 


Fig. 


14 


Fig. 


15 


Fig. 


16 


Fig. 


17 


Fig. 


18 


Fig. 


19 


Fig. 


20 


Fig. 


23 



EXPLANATION OF PLATES. 
Plate XXX. 

— Astraea sirius- Gould, animal from life. 

— Pyrene versicolor Sowerby var. atladona Duclos, animal from life. 

— Ogereulum of same. 

— Nerita melanotragus Smith, egg masses laid on a pebble. 

— A single egg mass of the above, enlarged. 

— Crepidula aculeata Gmelin, animal from life. 

— Chromodoris hennettii Angas, animal from life. 

— Rhinophore of same enlarged. 

— Branchiae of same enlarged. 

— Egg mass of same. 

Plate xxxi. 

— Argonauta hottgeri Maltzan. 

— Ceritliium dorsuosum Menke. 

— Hemidonax dactylus Hedley, type. 

— Pitaria inconstans Hedley, type. 

— Hinge of same. 

— Dorsal view of same enlarged. 

— Pseudarcopagia hotanioa Hedley. 

— Hinge of same. 

— Enlarged sculpture of same. 

Plate xxxii. 
, — Umhraculum hotanicum Hedley, type. 

Plate xxxiii. 
, — Tethys angasi Sowerby. 



317 



A REVISION OF THE AUSTRALIAN SPECIES OF THE GENUS 

BASSIA. 

By R. H. Anderson^ B.Se. (Agr.), Botanical Assistant. Botanic Gardens, 

Sydney. 

(Plates xxxiv.-xxxvi.) 

[Read 2oth July, i923.J 



Introductory. 

While working on some material from tiie Broken Hill District I was im- 
pressed by the state of confusion which apparently existed in the Family Cheno- 
podiaeeae. The exigencies of the moment forced me more i)articularly into the 
genus Bassia, and I quickly came to the conclusion that a complete revision of 
the Australian members of the genus was necessary. During the past eighteen 
months I have laboured to this end, and am now submitting a paper which, I 
trust, will do something towards straightening out a very chaotic genus. 

In this work I have been greatly assisted, in various ways, by the Director, 
Mr. J. H. Maiden, F.R.S., and by my fellow members of the staff, Messrs. E. 
Cheel, W. F. Blakely and J. N. Whittet. Mr. LaidlaAv, Government Botanist 
of Victoria, has greatly aided me by supplying, from the National Herbarium, 
types and other material, without which it would have been impossible to form 
definite decisions. Mr. C. T. White, Government Botanist of Queensland, was 
also kind enough to place ail his material at my disposal. Above all my thanks 
are due to Mr. J. M. Black of Adelaide who has assisted me in every way pos- 
sible. Yfe discovered that both of us were simultaneously working on the same 
genus, and, although Mr. Black had been studying the group for a long time, 
he co-operated with me in a very courteous and lielpful manner. I am in- 
debted to Miss E. King for the figures of the fruiting perianths included in the 
l^lates. 

I have, in the main, follovved Volkens's classification in Engler and Prantl 
(Pflanzenfamilien), but have added to and modified the various sections there 
arranged under Bassia. The great bulk of Australian material was not available 
to Volkens, whose classification, tlierefore, could only be provisional. 

I feel that, so far as species are concerned, the material at my disposal has 
been fairly thoroughly worked out, but the limitation of the genus and it- sec- 
tions is still a matter largely of opinion. 

Differences within the genus are at times so pronounced tliat one might be 



3.18 REVISION OF AUSTRALIAN SPECIES OF BASSIA^ 

inclined to split up the genus, making the sections distinct genera. It does not, 
however, seem advisable to follow this course, as the various species are obviously 
allied and connecting links between sections are not wanting. I have, however, 
excluded several species hitherto included in the genus. 

Historical. 

The genus Bassia was established in 1706 by Allioni (Melanges de philosophie 
et mathematique de la Societe Royale de Turin, iii., p. 177, t., 4). A fairly 
comi:>rehensive description was given, a translation of which will be found later 
in this paper (see p. 320), the type species, Bassia muricata All., being from 
Egypt. Five years later, in 1771, the name Bassia was also given to a genus 
in the Sapotaceae by Koenig (ex Linne Mant., ii., App. 555). 

The priority of Allioni's name is thus clearly established, but later botanists 
accepted Koenig-'s genus, placing Allioni's genus under subsequently described 
genera. The following genera in chronological order of establishment are now 
regarded as synonyms of the genus Bassia: — 

IVillemetia Maerkl., in Schrader, Joum. Bot., iii., 1, 1800, 329. 

Anisacantha R.Br. Prodr., 1810, 410. 

Sclerolaena R.Br. Prodr., 1810, 410. 

Villematia Moq., in Ann. sc. nat., 2 ser., i., 1834, 206, t. 6. 

Echinopsilon Moq., Ibid., ii., 1834, 127. 

Londesia Fiseh. et Mey. Index sem. Horti. petropoL, ii., 1835, 40, 

Kentropsis Moq. Chenopod. enum., 1840, 83. 

Maireana Moq. Chenopod. enum., 1840, 95. 

Centropsis Endl. Gen. Suppl., ii., 1842, 33. 

Eriochiton Y. Huell. Second Gen. Report, 1854, 15. 

Bissocarpm F. MuelL, in Trans. Phil. Instit. Victoria, ii., 1858, 75. 

Endlicher (Genera Plantarum, 1836-1840, 1927, p. 296), placed Bassia All. 
under Echinopsilon Moq. In 1849 De Candolle still retained Bassia All. under 
Echinopsilon Moq. and kept separate the genera Anisacantha and Sclerolaena 
established by Brown. In 1867 Bossier* made Bassia a section of Kochia Roth. 
In the Flora Australiensis (Vol. 5, 1870), Bentham made no recognition of 
Bassia All., but maintained the genera Anisacantha R.Br., Sclerolaena R.Br., and 
Chen^lea Thunb. Bentham and Hooker in 1880 (Genera Plantanim, iii., p. 60) 
placed Bassia All. under Chenolea Thunb. in the section Echinopsilon, maintain- 
ing Sclerolaena R.Br, and Anisacantha R.Br, as separate genera. 

In 1882 Mueller (First Census of Australian Plants) re-established the 
genus Bassia, uniting under it the genera Chenolea, Sclerolaena, Anisacantha, 
EcMnopsilon, Kentropsis, Dissocarpus, Eriochiton, Osteocarpum and Coilocarpiis. 
In the same year, he pointed out (Fragm., 12, p. 12) the claims of Allioni's 
genus to priority and re-established Illipe Koenig in the Sapotaceae, sinking 
Bassia Koenig as a synonym. 

In 1887 Sehweinfurth and Ascherson (Illustration de la Flore d'Egypt, p. 
127) likewise re-established the genus Bassia All., but giving it somewhat nar- 
rower limits than those assigned by Mueller. Prof. Baillon in the same year 
(Histoirc des Plantes Clienopndiares, 1877. 177) also recognised the genus. 

In 1880 Mueller (Vict, ^^at., v., 98) drew attention to the work of Drs. 
Ascherson and ScWp/einfuiih, and to that of Prof. Baillon. In the same year 



* I have not seen the work of Bossier but believe it is the "Flora Orientalis." 



BY R. H. ANDERSON. 319 

tile ''iconugTapliy of Australian Saisolaceous Plants" was published in wbieli 
Mueller figured the various species under Bassia. 

In 1891 Kuntze (Revisio Generum Plantarum, Part 2, p. 546) recognised the 
genus Bassia All., and also placed under it Chenolea Thunb. He also pointed 
out that if one would follow Bossier in his limitation of the genus, in which 
he placed Bassia as a section of Kochia, we would have to reverse the order and 
place Kochia as a section of Bassia. 

The genus Kochia was not established until 1800 by Eoth (Schrad. Journ. 
Bot., 1800 (1801), i., 307, t. 2), so Kuntze's contention is correct. He also 
drew attention to the work of Mueller in transferring all the Australian species 
to the genus Bassia. 

In 1893. Volkens (Pflanzenfamilien, iii. Teil, i. Halfte, Abteilung a, p. 
70) maintained the genus Bassia All. but kept Chenolea Thunb. as a .separate 
genus. He gave a short description of the genus and made 4 sections, namely, 
Echinopsilon, Anisacantha, Dissocarpus^ and Maireana. In the Index Kewensis 
(1893, p. 277) Bassia All, is given as a synonym of Chenolea Thunb. In the 
"Genera siphonogamarum" (1900-1907, p. 144) Torre and Harms gave a com- 
prehensive review of the genus and its synonyms, including the sections. 

In 1912 Muschier (Manual Flora of Egypt, i., p. 269), evidently following 
Bossier, placed the members of the genus Bassia under Kochia. But Bassia All. 
is now generally recognised by botanists as an established genus, although diiSer- 
ences of opinion exist as to its limitations. Merrill (Philippine Journ. Sc, 10, 
1915, p. 56) maintained the genus Bassia Koenig in the Sapotaceae, but was 
evidently unaware of the existence of Bassia All. Maiden and Betche (Census 
of N.S.W. PL, 1916, p. 68) follow the sections outlined by Volkens, the species 
being grouped accordingly. C. T. White (Q'ld. Agric. Journ., May, 1921, p. 
218) gives a brief survey of the genus under "Notes on the Genus Bassia." 

The genus Bassia All. is thus generally recognised by present day botanists 
as a well established genus, but its exact limitation is still a matter largely of 
opinion. It embraces species which appear to be generieally distinct, but be- 
tween extreme forms there are always connecting links, so that a satisfactory 
splitting up of the genus (would be very difficult. The section Maireana might 
perhaps be given generic rank as a genus intermediate between Kochia and 
Bassia. 

Family CHENOPODIACEAE. 

Division Ci/clolobeae-Camphorosmeae. 

Bassia AIL, Misc. taurin., iii. (1766), 177, t. 4. 

Flowers hermaphrodite. Perianth hirsute, rarely glabrous, urceolate, globu- 
lar, or depressed, with 4-5 short, inflexed or erect lobes, usually hardened but 
sometimes membraneous on the fruit and with 2 or more erect, outwardly spread- 
ing, or recurved spines. Styles 2-3, connate at the base, thread-like. Stamens 
usually 5, occasionally fewer. Fruit enclosed in the perianth, ovoid, globular or 
depressed. Pericarp membraneous. Seed globular or compressed, vertical, hori- 
zontal or oblique. Embryo annular or almost so, surrounding a mealy albumen; 
radicle usually erect or ascending. Testa membraneous. Undershrubs, shrubs, 
or occasionally annuals, prostrate, decumbent or lerect, usually divaricately 
branched. Leaves alternate, rarely opposite, usually linear or lanceolate but in 
some species very broadly laiiceolat-e to rhomboid. Flowers solitary or in 
clusters, usually in the axils of the leaves. Fruiting perianth usually closely 
sessile. 



320 REMSION or AUSTRALIAX SPECIKS OF BASSIA^ 

The species are distributed in Middle Europe, North Africa, Temperate 
Asia, but most abundantly in Australia. In this revision 42 species are re- 
corded for Australia, the majority of which are to be found under dry arid con- 
ditions. A few species, however, occur in more temperate localities and on 
richer soil. 

The following is a translation of the original description Ijy AUioni wliieh 
is very difficult to obtain. 

"This new genus of plants for which I have instituted the name Bassia, has 
a terete, branching, fibrous, perennial root, a round stem, not very stout, and 
branching continuously and intricately, the branches branching in like manner 
again. The first branches of the stem are almost decumbent. The v/hole plant is 
villous hirsute, equal in height to 1-2 spans. The leaves are linear lanceolate, 
thick, compressed, sessile, alternate, or rather densely and irregularly placed, 
usually of 2 kinds, one of which is longer than the other, and with 2 sessile flowers 
present in its axil. The other is shorter, with one sessile flower, and with a bundle 
of leaves making a new branch. All flowers are sessile in the axils of the leaves. 
The flower is apetalous, calyx monophyllous, five times divided into acute segments. 
Stamens five, bearing the yellow didymous anthers above the calyx. Styles 2, 
filiform, from white to purple, arising from the embryo. When in flower, however, 
the flat coalescing segments of the calyx enclose the flower, and the calyx be- 
comes the capsule, containing when mature the orbicular seed of indistinct colour, 
not cochleate, but having on the upper surface a circular engraved groove. In 
that portion in which the embryo is of larger size, the receptaculum is extended 
to greater dimensions, and at the base of the calyx 5 yellov.'ish spines appear, 
situated at equal distances from one another. The calyx, becoming thickened, 
firm, and coriaceous, forms the rounded flattened fruit, having 5 firm spines 
radiating from its circumference. This genus lies between Chenopodium and 
SalsC'la. But from both it differs, especially in the changing of the calyx to the 
capsule. On account of this it seems to be worthy of establishment as a genus, 
and I have named it Bassia as a token of my friendship, gratitude, and esteem, 
for the most learned Ferdinand Bassium of Bonn." 

In dividing the geims into sections I have adopted the 4 sections of 
Volkens, but have established two additional ones. 

Section I. Echinopsilon Hook, f., in Bentham and Hooker, Grcnera Plant., 
lii., 1880, 60 (Sect. Chenoleae) ; Volkens, Engler et PrantL, Pfizfam,, iii., ia., 
1893, 70. 

This section includes Willemetia Maerkl., Villemetia Moq., Ecliinopsilon Mo(j., 
Londesia Fiseh. et Mey. 

Floral envelope remaining membraneous or very slightly hardened, the 
spines (occasionally reduced to small protuberances) arising on the top of the 
free lol>es of the floral envelope, or directly under these. In this section are 
included mainly the Arabian and Egyptian species.* 

After some hesitation I decided to place in this section tlie four Australian 
species, B. demiflora W. V. Fitzg., B. Muelleri (Benth.) F.v.M., B. eiirotioides 
F.V.M., and B. astrocarpa F.v.M. in which the perianth is only slightly or not 
at all hardened. The other species, e.g., B. Tatei F.v.M. and B. eriacunthn 
(F.v.M.), in which the perianth is only slightly hardened, 1 liave placed in 
Section HI. {Anisacmxtha) , but they might perhaps be gathered together, leav- 
ing only the species with much hardened perianths in Section Anisacantha. 

The Egyptian spe(-ies B. muricata All., whicli Volkens quotes as an example 
in the section Eehinopsilou (Engler and PrantL, P/lzfam., iii., ia., 70), has 
certainly a sliglitly hardened base and perianth. 

Section II. Eriociiiton. 

I am proposing lliis as a new section. It resembles Sect. Ecliinopsilon, but 



BY B. £1. AXDERSO-N. 321 

the perianth, in addition to the spines, bears 5 dorsal erect membraneous ap- 
pendages, resembling, in this respect, some species of Koehia. 

Tlie section includes the genus Eriochiton F.v.M. (Sec. Gen. Kept., 1854, 15) 
and is represented l)y a single species, B. selerolaenoides F.v.M. 

SectiOX 111. Anisac.vxtha Volkens in Engler and Prantl, Ptlzfaui., iii.. ia., 
1893, 70. 

This section includes Aulsacantha R.Br., Sclerolaena E.Br.. Kentropsis Moq. 
and Ceniropsis Endl. 

The floral envelope hardens at the base, developing witli the fruit, forming 
a hard fruiting perianth, bearing llic spines. Seed vertical, horizontal or 
oblique. 

This section includes the gTeat majority of the Australian species. 

Seotiox IV. DissocARPTJS Volkens, Engler and Prantl. Pflzfam.. iii., ia., 
1893, 70. 

This section includes Dissocarpus F.v.M. The flowei-s grow together in a 
cluster, the fruits subsecjuently forming a hard mass of 2 to many united in- 
dividuals. 

This section includes 3 Australian species. B. hi flora (E.Br.) F.v.M.. B. 
paradoxa (R.Br.) F.v.M. and B. spinosa Ewart and Davies. 

Section V. Platyacantha F.v.M., Frag-m., 12, 1882, p. 12. 

In his description of Bassia tridens F.v.M., Mueller Kvrites: — "The species 
merits a separate section under the name of PlatijacantJia^ unless it remains with 
the preceding species (B. Forrestiana) in Anisaeantha." Fruiting perianth with 
broad flattened spine-like appendages, both spines and perianth hardened. 

The section is quite a distinct one, and contains only one species, Bassia 
tridens F.v.M. It approaches the next section, Maireana, in some respects, but 
differs in the regular well-defined spines, and the absence of the horizontal ex- 
pansion. 

Section VI. Maireana Volkens, Engler and Prantl, Pflzfam., iii., ia., 1893, 
70. 

This section includes Maireana Moq. It resembles Section Anisaeantha but 
the hardened fruiting perianth develops n small wing-like border or horizontal 
expansion, the edge of which is broken up into spine.s. A connecting link with 
Koehia. 

This section includes 3 Australian species. B. stelligera F.v.M., B. Luelr- 
manni F.v.M.. and B. hracln/ptera (F.v.M.) nov. comb. B. microcarpa, n.sp., 
also lias certain affinities with this section. 

7<^e}i to the Sections. 
1. Perianth remaining membraneous, not at all, or very slightly hardened in the 
fruiting stage. 

2. Fruiting perianth with erect membraneous appendages in addition to the 
spines Sect. ii. Eriochiton. 

2a. Fruiting perianth without any appendages other than the spines 

Sect. i. Ecliinopsilon. 
la. Perianth becoming hardened from the base, tistially much hardened in fruit. 

3. Flowers clustered, the fruiting perianths connate into a hard mass of 
two to' many fruilia Sect. iv. Dissocarpus. 

3a. Flowers solitary, the fruiting perianths not connate. 

4. Fruiting perianth bordered by a narrow rim or wing-like expansion, 

broken up into spines Sect. vi. Maireana. 

4a. Fruiting perianth without any wing-like expansion, spines present. 



322 REVISION^ OF AUSTRALIAN SPECIES OP BASSIA, 

5. Spines broadly flattened, forming horizontal appendages 

Sect. V. Platyacantha. 
5a. Spines net flattened, acicular Sect. iii. Anisacantha. 

Section I. Echinopsilon. 

1. Fruiting perianth covered in a short dense tomentum. never silky hairy. 

Spines 5, much shorter than the tube. A. B. astrocarpa F.v.M. 

la. Fruiting perianth with a vestiture of long silky hairs. Spines usually as long 
or longer than the tube. 

2. Spines stout, usually under 3 mm. long. .. 3. B. MiieUeri (Benth.) F.v.M. 
2a. Spines slender, or reduced to awns, more than 3 mm. long. 

3. Fruiting perianth clothed in long silky hairs which are up to 10 mm. 
long. Spines reduced to 4-6 thread-like awns, hardening towards the 

base, from 8-15 mm. long 1. B. euroiioides IF.v.M: 

3a. Fruiting perianth with short silky hairs, rarely 5 mm. long. Spines 
slender, under 7 mm. long, the fruits very dense towards the end of 
the branches 2. B. densiflora W. Y. Fitz. 

Section 2. Eriochiton^. 

Fruiting perianth enveloped in wool, with 5 erect, bifid, membraneous appendages, 
and with 5 horizontally spreading, slender spines. 5. B. sclerolaenoides E.v.M. 

Section 3. Anisacantha. 
Kei/ to the Series. 

1. Fruiting perianth much hollowed at base, the cavity occupying almost or more 

than half the perianth. Seed vertical, horizontal, or oblique Series I. 

la. Fruiting perianth not at all, or only slightly, hollowed at base, the cavity, if 

any, occupying much less than half the perianth. 

2. Seed vertical or vertically oblique. (Sometimes almost horizontal in 
B. tricuspis (F.v.M.) and B. quinqueci(~spis F.v.M Series 2. 

2a. 'Seed horizontal or horizontally oblique. Series 3. 

Series 1. 

1. Fruits enveloped in long silky hairs. Spines 2. or very occasionally 3 

6. B. eriacantlia (F.v.M.), ii. comh. 
la. Fruits shortly tomentose or glabrous. 

2. Fruits with 5-6 spines, glabrous. Seed horizontal 

8. B. anisaccmtlwides (F.v.M.), n. comb. 
2a. Fruits with 2-3 spines. 

3. iSeed horizontal (or very occasionally vertical ?). Fruiting perianth 
with 2, more or less equal, divaricate spines, with a small tubercle at 

the base of one 7. B. uniflora (R.Br.) F.v.M. 

3a. Seed vertical or oblique. Base of fruiting perianth swollen or dis- 
tended. 

4. Fruiting perianth soft in texture, usually ribbed or furrowed, spines 
2, irregular and divaricate, with usually a third spine on the base 
of one. Often only one spine developed. Leaves broad lanceolate, 

or lanceolate g j^ rj.^^^,- y.vM. 

4a. Fruitmg perianth hard, scarcely ribbed or furrowed, with 2 fine, 
equal, parallel spines, very rarely divaricate. Leaves linear. 

10. B. paraJlelic>isj)/.s, ii.sp. 



BY R, IT, ANDERSON. 



323 



Series 2. 
1. Spines of the fruiting perianth 2-4 in number. (Sometimes 5-spined in 
B. decurrens. 
2. Whole plant, except the flowers, glabrous or almost so. 

3. Longest spine more than 15 mm. and up to 35 mm. long. Usually 4 
spines, 2 of which are more or less reduced. 14. B. longicuspis F.v.M. 
3a. Longest spine usually less than 15 mm. long. 

4. Spines usually 2 in number, more or less equal, or with 1-2 shorter 
spines or tubercles. 
5. All spines very short, the longest less than 5 mm. long. 

6. Spines usually recurved, 4 in number, the longer pair op- 
posite and diverging, the shorter pair approximating each 
other at their base 11. B. recurvicuspis W. V. Fitz. 

6a. Spines not recurved, usually 2 in number, with a 3rd very 

short spine or tubercle 12. B. glabra F.v.M. 

5a. Longest spine more than 5 mm. and up to 15 mm. long. 

7. Limb twice as long as the tube of perianth. Spines 2, with 
a short tubercle, or with a 3rd small spine . . . 

13. B. hicus'pis F.v.M. 

7a. Limb not as long as tube. Spines 3-4 

18. B. Drummondii (Benth.) F.v.M. 

4a. Spines 3, rarely 4, more or less equal, or with a fourth very short 
one. 
8. Fruiting perianth attached by a very oblique base. Spines 3 

with a fourth much reduced. Limb recurved 

15. B. divaricata (R.Br.) F.v.M. 

8a. Fruiting perianth attached by a broad distended base. Spines 

3 or very occasionally 4, all more or less eqtial. Limb erect. . . 

16. B. tricu>ipis {F.v.M.), n. comb. 
2a. Plants pubescent, hirsute or tomentose. 

9. Fruiting perianth enveloped in long silky hairs 

17. B. lamcuspis F.v.M. 
9a. Fruiting perianth tomentose, shortly hirsute or almost glabrous. 

10. Limb equal to the length of tube or almost so. Spines in pairs 
with 1-3 much smaller ones decurrent o'n one spine. Tube pro- 
minently ribbed 19. B. deciirrens J. M. Black. 

10a. Limb much shorter than the tube. 

11. Fruiting perianth with 3-4 spines, 2 of which are more or less 

equal, the others smaller 

18. B. Drummondii (Benth.) F.v.M. 
11a. Fruiting perianth with 2 spines, visually with a small tubercle 
at the base of one. 

12. Spines of the fruiting perianth spreading outwards in 
the same vertical plane. Limb fairly well produced. .. 

21. B. patenticuspis, n.sp. 
12a. Spines of the fruiting perianth spreading outwards but 

twisted in different directions, not in the same vertical 
plane. Limb much reduced. Base of fruit much furrowed. 

20. B. nhliquicHi^pifi, n.sp. 

la. Spines of the fruiting perianth 5-6 in number, or 4 with one spine bifid. 

13. Spines 6. or 5, with the fifth bifid. Fruiting perianth oblong, small, rarely 

more than 3 mm. long, ribbed or striate and produced at the base into 2 

short spurs. Leaves usually short, linear, glabrous or hirsute 

22. B. ecMnopsila F.v.M. 
13a. Spines 5, or 4, with the fourth bifid. Fruiting perianth cylindrical or de- 
pressed, without any spurs at the base. 



324 REVISION OF AUSTRALIAN SPP:CIES OP BASSIA^ 

14. Fruiting perianth depressed, irregular in shape, or shortly cylindrical, 
the tube rarely more than 2 mm. long. 25. B. qidinquecuspis F.v.M. 
14a. Fruiting perianth cylindrical or conical, the tube usually more than 
3 mm. long. 

15. Base of mature perianth hollowed, slightly oblique. Leaves 
and branches densely or sparingly hirsute or tomentose. 

Branches slender, ascending 23. B. tubata, n.sp. 

15a. Base of mature perianth not hollowed, very oblique. Leaves 
and branches glabrous, except for tufts of hairs in the axils of 
the leaves. Branches very intricate. .. 24. B. intricata, n.sp. 

Series 3. 
1. Spines of the fruiting perianth 2-4 in number. (Rarely 5-spined in 

B. ventricosa and B. Forrestiana). 
2. Fruiting perianth always 3^ — 4-spined (or occasionally 5-spined in B. ven- 
tricose and B- Forrestiana) , hirsute or tomentose. 

3. Longest spines of the fruiting perianth more than three times the 
length of the tube and up to 25 mm. long. Leaves densely hirsute 
and up to 3 mm. long 26. B. Forrestiana F.v.M. 

3a. Longest spine of the fruiting perianth barely twice as long as the 
tube, rarely exceeding 6 mm. Leaves hirsute, but becoming almost 

glabrous. 27. B. ventricosa J. M. Bla<?k. 

2a. Fruiting perianth usually 2-spined, but often with a third spine on one 
. face, densely woolty tomentose. 

4, Tube rareh^ more than 3 mm. long, the limb almost or quite equal to 
the length of the tube. Whole plant always hirsute or tomentose. 

28. B. limbafa J. M. Black. 
4a. Tube more than 3 mm. long, the limb usually much shorter than the 

tube. Older leaves often becoming glabrous or almost so 

29. B. hicornis (Lindl.) F.v.M. 
la. Spines of the fruiting perianth 5-6, or 4 with one spine bifid. 

5. Spines of the fruiting perianth usually 6, or 5 with one spine bifid. 

6. Fruiting perianth large, always more than 3 mm. and up to 6 mm. m 
diameter at the top. Spines up to 6 mm. long. Plant woolly tomen- 
tose 30. B. Cornishiana F.v.M. 

6a. Fruiting perianth small, under 3 mm. in diameter, the spines ^-3 mm. 
long. Plant denseh^ or sparingly hirsute. 

7. Leaves invested with dense long hairs. Fruiting perianth very 
."■nTiall. the longer spines under 1 mm. long and sharply hooked at 
the end 35. B. microcarpa, n..sp. 

33. B. parviflora, n.sp. 
7a. Leaves sparingly hirsute. Fruiting perianth small, the longer 

spines usualh^ exceeding 1 mm. in length and not hooked 

5a. Spines of the fruiting perianth 5, or 4 with one spine bifid. 

8. Fruiting perianth readily detached from the stem, the tube 
narrowing to a regular circular base. 

9. Fruiting perianth tomentose, the top domed, the spines some- 
what ascending .32. B. convexuU, n.sp. 

9a. Fruiting perianth glabrous or slightly hirsute, the top fiat, 
spines horizontal or slightly recurved- Tube strongly ribbed. 

34. B. costata, n.sp. 
8a. Fruiting perianth detached with difficulty from the stem, the tube 
very irregular, the base usually oblic|ue. 

10. Plant glabrous or villose, scarcely robust. Leaves linear, 

lanceolate, or soinewhat obovate 

25. Ji. qiiinqiM'cuspif! F.v.M. 



By K. 11. ANDERSON. 325 

10a. Plant woolly tomentose, robust. Leaves obovate. 

31. B. Birchii F.v.M. 

Section IV. Dissocarpus, 

1. Connate fruiting perianths always furnished with spines, usually very spiny. 
(Occasionally the spines reduced to obtuse horns in B. paradoxa). 

2. Fruiting perianths 6 or more in the head, the spines simple 

37. B. paradoxa (R.Br.) F.n.M. 
2a. Fruiting perianths less than 6 in the head, the spines stronger and usually 

divided 38. B. spinosa Ewart & Daviefe. 

la. Connate fruiting perianths usually without spines, or with 1-3 short spines or 
tubercles. Fruiting perianths 2-6 in the head. 36. B. bifk>ra (R.Bv.) F.v.M. 

Section V. Plattacantha. 

Spines 3, with one bifid, much flattened, and somewhat united at their base. Base 
of tube usvtall}'- hollow . 39. B. tridens F.v.M. 

Section VI. Maireana. 

1. Horizontal expansion of the fruiting perianth irregularly broken up into spines 
or spine Hke segments. 
2. Tube globose or globose truncate, more than 1 mm. in diameter. Leaves 

linear 41. B. steUigera F.v.M. 

2a. Tube very short and almost flattened, less than 1 mm. long. Leaves 

rhomboidal, tapering into a long narrow petiole-like base 

40. B. LueJimamii F.v.M. 
la. Horizontal expansion not broken up but marked by 5 regularly placed, verj' 

short spines. Leaves with long dense hairs 

42. B. hrachyptera (F.v.M.), n. comb. 
1. Bassia eurotioides F.v.M. 
EcJdnopsiion eurotioides, F.v.M., Fragm., vii., 1869, 13. — CJiowlea eurotioides, 
F.v.M., B.F].. y., 1870, 191.— 5. eurotioides, First Census, 1882; Vict. Nat., v., 
1889, 98; Icon. Austr. Salsol.. Plate Ixxxi. 

Notes additional to the description. — Spines reduced to 3-6 slender awns, 
tapering from a rather broad and somewhat thickened base. Perianth domed. 
Awns up to 15 mm. long. Vestiture on older branches considerably reduced. 

Affinitif. — This species sho'ws closest affinity Avith B. densiflora W. V. Fitz. 
which differs in the shorter rigid spines, and the sliorter vestiture on the fruiting 
periaJith. 

Range. — So far as we know it is confined to Western Australia. The fol- 
lowing localities are represented in the National Herbarium: — Nannine, Boulder, 
Comet Vale, and between Kunnunoppin and Mt. Marshall, and Lake Bai-lee. 

2. Bassia dei^siflora W. V. Fitz. 

Mueller Bot. Soc, 1904, 31. 

Notes additioncd to the description. — A low spreading shrub densely invested 
with wliite silky hairs. Perianth lobes usually 4, forming a square shaped top 
to the fruiting perianth and each bearing at the comer a spine, one of which 
is generally bififl. Spines slender, more or less equal, up to 6 mm. long, oc- 
casionally 6 in immber. Leaves clothed in longer hairs (up to 15 mm.) than 
the fruiting perianth. Branches at times only sparingly hirsute. 

Affinities. — 1. With B. eurotioides (Sec previous species). 2. In its 



326 REVISION CF AUSTRALIAN SPECIES OP BASSIA. 

general superficial appearance it resembles B. lanicuspis, but differs very con- 
siderably in the nature of the fruiting iierianth. 

Bmige. — Apparently confined to Western Australia. The type specimens 
are from Nannine (W. V. Fitzgerald 9/1903) and Gwalia (W. V. Fitzgerald 
11/1903). Both of these are represented in the National Herbarium, and in 
addition specimens are present from Millevva, Cue, Laverton, and Cowcowing 
(Mixed Avith B. eurotioides F.v.M.). 

3. Bassia Muelleri (Beuth.) F.v.M. 

Chenolea Muelleri, Benth., B.FL, v., 1870, 191, — B. Muelleri, First Census, 
1882; Fragm., xii., 1882, 13. 

The only specimen I have seen belonging to this species is the type from 
Sturt's Creek (Mueller). 

The infiorescence and leaves are clothed in long silky hairs, the vestiture on 
the branches being much shorter. Leaves up to 15 mm. long by 14-2 mm. wide. 
Spine. 4-5, short, stout, the longest 2-2J mm. long. Perianth very small, barely 
2 mm. long. The styles appear to be more commonly 3 in the specimen ex- 
amined — Bentham gives the styles as 2 (B.FL, v., 191). The seed, though not 
mature, is more inclined to the vertical than the horizontal, although Bentham 
describes it as horizontal or oblique. 

I have placed this species in the Section Echinopsilon as the perianth ap- 
parently remains membraneous, but mature fruits have not yet been found. 

It is an imperfectly known species, and more material is necessary for its 
satisfactory definition. 

W. V. Fitzgerald has recorded it from Broome, W.A. (Bot. of the Kimber- 
leys, Journ. Proe. Roy. Soe. W.A., iii., 1918, 36) . 

4. Bassia astrooarpa F.v.M. 

Chenolea astrocarpu, F.v.M., coll. 1 B. astrancantha F.v.M., First Census, 
1882, 140.— B. astrocarpa, Fragm., xii., 1882, 12; Second Census, 1889; Icon. 
Austr. Salsol., Plate Ixv. 

The following is a translation of the original description : — ''Silky-tomen- 
tose, leaves somewhat short, semiterete-linear or the upper ones more lanceolate, 
acute, scarcely naiTOwed at the base. Spines of the fruiting calyx 5, con- 
spicuously shorter than the tube, slightlj' unequal, tomentose, pubescent under- 
neath, the tube about twice as long as broad, slightly hollowed at the base, the 
seed vertical. In the neighbourhood of Nichol Bay, Crouch. A somewhat 
harsh shrub. Leaves often between 3 and 5 lines long, intermingled towards 
the tops of the branches with the fruits, often closely crowded. Tube of fruit- 
ing calyx about 2 lines long, spines diverging almost in a stellate fashion, 
scarcely exceeding, or shorter than a line. Approaches B. MiMlleri, differing in 
the more lax vestiture, the crowded flowers, })ut chiefly in the length of the 
fruiting calyx, and perliaps in the position of the seed. But I liave not the 
mature fruit f>f B. Muelleri. St^'les of this plant often 3." (F.v.M., Fragm., 
xii., 12). 

Additional notes. — Spinels 2-3 ram. long. Tube of the fruiting perianth up 
to 5 mm. long by 2 mm. in width. Leaves up to 15 mm. long. A species 
readily distinguishable by the long tnl)(! and IIk' 5 short spines. 



BY K. 11. AXUERSON. 327 

Mueller included it in the section Anisacantha. but tbe scarcely hardened 
periantli seems to bring it into EcMnopsilon. 

Ajfinity. — This species approaches B. recurvicuspis W. V. Fitz., but differs 
in the softer character of the perianth, and the number and nature of the 
spines. 

Range. — Apparently limited to Western Australia. Specimens from Broome 
and Goodygoody are included in the National Herbarium. 

5. Bassia sclerolaenoides F.v.M. 

Eriocliiton sclerolaenokles F.v.M., Second Report, 1854, 15. — Echinopsilon 
sclerolaenoides F.v.M., Trans. Phil. Inst. Vic, ii., 1858, 75; Fragm., vii., 1869, 
13. — Chenolea sclerolaenoides F.v.M., in B.FL, v., 1870, 192, — Chenolea Dcdlach- 
yana Benth., FL, v., 1870, 191. — Bassia Ballacliyana (Benth.), F.v.M., First 
Census, 1882.— B. sclerolaenoides, First Census, 1882; Fragiu., xii., 1882, 12; 
Icon. Austr. SalsoL, PI. Ixxxii. — Bassia eriochiton Tate, Flora S.A., 1890, 51. 

Notes additional to the description in B.FL, v., 192. — Fruiting perianth en- 
veloped in dense wool which often conceals the spines. Spines up to 7 mm. 
long. Leaves up to 15 mm. long but usually much shoi'ter. Styles usually 2, 
connate at the base, but occasionally 3, or the third much reduced. 

This species is quite distinct from other members of the genus by reason 
of the erect membraneous appendages. These are also found in some species of 
Kocliia. The scarcely hardened perianth places it in the Echinopsilon section, 
but the membraneous appendages warrant *it being made a separate section. 

Synonym. — B. Dallachyana (Benth.) F.v.M., First Census, 1882. Chenolea 
Dallachyana Benth., Fl., v., 1870, 191. 

A considerable amount of confusion has existed as regards the above species. 
The material in the National Herbarium labelled B. Dallachyana proved to be 
Kochia eriantha F.v.M. or an allied species of Kochia, and the confusion ap- 
parently existed in other States as well. 

All the material had 'well developed wing expansion, usually irregularly 
separated into 5 lobes or segments, the whole fruiting perianth being concealed 
under long dense hairs. Such material has no affinity either with the descrip- 
tion or type specimen of B. Dallachyana (Benth.) F.v.M. 

A close examination of the type specimen (Murray Desert, Dallachy) shows 
that it can hardly be separated from B. sclerolaenoides F.v.M., and IMueller in 
the unpublished manuscript of his Third Census considered it a synonym of 
B. sclerolaenoides F.v.M. 

Bentham in his key (B. Fl., v., 1870), distinguishes the two species by the 
absence of appendages to the fruiting perianth of B. Dallachyana, but remarks 
that the fruiting perianth may not be cjuite ripe. 

The perianths of the type were very immature, as in no case could I find 
mature, or even immature seeds developed. The flowers, vestiture, and general 
habit closely resembled those of B. sclerolaenoides F.v.M. and the absence of the 
appendages might possibly be due to the undeveloped condition of the perianth. 

It would seem that, at present, there is insufficient evidence to distinguish 
the two species. 

Should additional material show that the appendages are always wanting in 
B. Dallachyana, then the species would have to be transferred from the genus 
Bassia, probably to the genus Chenolea Thuub. 

Range. — Found in arid parts of Western Australia, South Australia. Vie- 



328 REA-ISION Oi" AUSTRALIAX SPECIES OF BASSIA^ 

toria and New South Wales. The following localities are represented in the 
National Herbarium: — Western Australia: Coolgardie, Nannine, Victoria Desert 
Camp 53. South Australia: MuiTay River, Port Pirie, Mount Lyndhurst, Moo- 
looloo Station. Netv Sotith Wales: Bi'oken Hill, Barrier Ranges. Victoria: 
Ouyen. 

G. Bassia EPaACAXTHA (F.V.M.), n.comb. 

Kentropsis eriacantJm F.v.M., Fragm., ii., 140. — Sclerolaena lanicuspis 
F.V.M., B.FL, v., 1870, 195. 

A remarkable confusion has existed in regard to this species. Mueller (I.e.) 
described a specimen from the Darling River as Kentropsis eriacantha. At page 
170 of the same work he redescribes the si^ecies as AnisaeantJia lanicuspis, giving 
as type sj^ecimens the Darling River specimen, previously described as Ken- 
tropsis criacontha, and also a specimen from the Barrier Ranges. Kentropsis 
eriacamtha is placed as a synonj'm of Anisacantha lanicuspis (Fragxa., vii., 14)1 
and, subsequently, this was followed in other works and by other Hvriters. An 
examination of the material at present under B. lanicuspis, however, shows that 
there are two distinct species present. For convenience I will term these A. and 
B. 

Species A. has larger fruits and leaves than species B., is occasionally 3-, 
but usually 2-spined, the leaves are al'ways densely hirsute, the fruiting perianth 
much hollowed at the 1)ase, and the seed obliquel}', or almost horizontally, placed. 
Species B. has smaller fruits and leaves, is 2 — 4-spined, the leaves tend to 
lose much of their vestiture and the fruiting perianth is oblong or turbinate, 
with a scarcely hollo^ived base. The seed is vertical. The shape and base of the 
fniiting perianth is qviite diiferent from that of species A. 

The two species resemble each other in the long silky vestiture of the 
flowers and fruits. 

The species B. is apparently the more common of the two. Mueller (Icon. 
Austr. Salsol., Plate Ixxx.) has apparently figured the fruiting" perianths of both 
species. Fig. 6 represents species A. and Fig. 5 appears to be species B. 

On obtaining the types I found that tlie Darling River specimen, originally 
described as Kentropsis eriacantha (Fragm., ii., 140) agreed with the species 
A. I therefore intend taking up the specific name eriacantlia, and reconstituting 
the species under Bassia. 

In regard to Anisacantha lanicuspis, this species was regarded by ]MucxIer 
as the same species as Kentropis eriacantha, and he quotes the Darling River 
specimen as one of the types of the former species. The other type, however, 
from tlje Barrier Ranges, belongs to tlie species B. and it was on this specimen, 
appaix-ntly, that IMueller drew up his description of Anisaenntho lamcnspis. 

1 therefoi-e intend maintaining the name B. lanicuspis, but limiting it strict- 
ly to the species B. 

The Sclerolaena lanicuspis described by Bentham (B.FL, v., 195) is B. 
eriacantha. His deserii»tion applies to this species and not to Amsacantha lani- 
cuspis F.V.M., and he also quotes the Darling R. specimen. 

Notes on B. eriacantha (F.v.M.). — It is usually a small shrul) or under- 
shrub, erect, and clothed in a dense silky vestiture. The leaves are thick and 
up to 30 mm. long. Tlie fniiting perianth is densely hirsute, usually 2- but 
ocea-sionally 3-spined. witli a large hollow base. Spines up to 9 mm. long. 
Reed obliquely placed. l)iit inclined to the horizontal. 

lianrje. — Tt occurs in Western .Nustrali.-i, Soutli Australia. ;hh1 New South 



BY R. ir. ANDERSON. 329 

Wales. The following localities are represented in the National Herbarium:— 
W.A.: Cue, Nauziine; N.S.IV.: Umberuniberka, Barrier Ranges. 

7. Bassia uxiflora (K.Br.) l\v.M. 

Sclerolaena uiti flora R.Br., Prodr., 1810, 410 and in B.FL, v.., 1870, 194. — 
Anisacantha diacantha Nees, PI. Preiss., i., 635. — Kentropsis diacantha Moq., 
D.C., Prodr., siii., ii., 138. — Anisacantha kentropsidea F.v.M., Trans. Vict. Inst., 
1855, 133; Hook. Kew Journ., viii., 204 (reduced to A. diacantha in Fragin., vii., 
14). — Sclerolaena diacantha Benth, B.FL, v., 1870, 194. — B. uniflora, F.v.M., 
First Census, 1882. 

A very variable species but always tomentose or hirsute. Leaves from. 
short, thick, and obtuse, to long, linear, and almost acute and up .to 30 mm. 
long. Spines 2, usually with a tuljercle at the base of one, up to 7 mm. long, 
but at times reduced to veiy short spines or tubercles. Base of fniiting perianth 
much hollowed and showing radiating septa. 

Hitherto the two species, B. uniflora (R.Br.) F.v.M. and B. diacantha (Nees) 
F.v.M., have been kept separate although both Mueller and Bentham recognised 
their clovse aflinity. Bentliam (B.FL, v., 194), in a foot-note to Sclerolaena %mi- 
flora, suggests that S. diacantha may indeed prove to be a variety of S. uniflora. 
Mueller (Viet. Nat., vii., 66) states that B. uniflora is "only a variety of B. 
diacantha." The older name (i.e., uniflora), however, would have to be adopted. 

In his key, Bentham separated the two species on the obtuse leaves and short 
spines of the one as opposed to the almost acute leaves and longer spines of the 
other. 

An examination of the material shows that there are two extreme types, 
namely, one with thick obtuse leaves and very short spines {uniflora type), and 
the other Avith long, linear, rather acute leaves and longer spines [diacantha 
type). The former is generally found in coastal regions, whereas the latter is 
usually some distance inland. But it is quite impossible to separate the material 
along these lines. We have specimens with the thick obtuse leaves but long 
spines; others 'with the long linear leaves but very short spines; and all manner 
of intermediate forms. In addition, the uniflora type is at times found inland 
(Ta-rcoola, J. M. Black) . Thus we cannot sepai'ate the species either as regards 
length of spines, type of leaves, or locality, and for the same reason it is im- 
possible to make a well defined variet}', although the two extremes of the species 
look veiy different. Generally speaking the uniflora type occurs on the coast and 
gradually turns into the diacantha type as we go inland from the sea- 

I have been unable to see the type specimen of Anisacantha kentropsidea 
F.v.M., which both Mueller and Bentham regarded as a synonym of Anisacantha 
diacantha, but am content to let it remain as such. 

Sclerolaena diacantha var. longispinea ? Benth. (B.FL, v., 195). 

Under this variety Bentham included two distinct species which I am describ- 
ing as B. patenticuspis and B. ohliquicuspis (see under B. ohliquicuspis). 

Mr. J. M. Black has drawn my attention to what appears to be a form of 
B. uniflora ■\vith a vertically placed seed. He sent over tAvo South Australian 
specimens (Hergott and Arkianga Creek) and there is, in the National Herljarium, 
a similar specimen from Mr. Lyndhurst. These specimens differ from typical 
B. uniflora in the position of the seed, which is vertical, whereas in typical B. 
imi flora I have always found it to be horizontal. In general appearance, the 



330 REVISION OF AUSTRALIAN SPECIES OF BASSIA_, 

specimens are identical with B. uniflora but differ slightly in the fruiting perianth 
and spines. The tube is longer and more constricted above the dilated hollow 
base than in typical B. uniflora, thus making room for a vertical seed. The 
opines are also more unequal in length. Although resembling B. parallelicu-spis, 
n.sp., in the position of the seed, the specimens are readily separable from that 
species by the shape of the fruiting perianth and the stouter, divaricate, unequal 
spines. 

Although these specimens show a decided variation from typical B. uniflora 
in the position of the seed, and minor differences in the tube and spine, they ap- 
pear to belong to that species and I hesitate from distinguishing them even as a 
variety. Additional specimens and further field observations would serve to 
throw light on this point. 

There is also in the National Herbarium a specimen from Mt. Sturt Station 
(A. Morris) which is distinguished from the more typical forms of B. uniflora 
(R.Br.) r.v.M. by the long thick leaves vvhich are up to 40 mm. in length. The 
spines of the fruiting perianth are more woody than those of typical B. uniflora, 
but the specimen appears to be only a form of a very variable species. 

Bange. — B. uniflora has a fairly extensive range in the drier parts of Western 
Australia, South Australia, Victoria and New South Wales, a large number of 
loeaKties being represented in the National Herbarium. The extreme uniflora 
type is represented by specimens from St. Francis IsL, Nuyt's Archipelago, Murat 
Bay, Tarcoola. 

The specimen determined by Mueller from the Elder Exploring Expedition 
(Trans. Roy. Soc. S.A., xvi., 1892, 346), Victoria Desert Camp 53, is an inter- 
mediate form between the uniflora and diacantha forms. 

The diacantJia form is much the more common. 

8. Bassia ANiSACANTHOiDES (F.v.M.) , n. comb. 

Echinopsilon anisacanthoides F.v.M., Trans. Phil. Instit. Viet., ii., 1858, 76. — 
Anisacantha hrevicuspis F.v.M., Fragm., iv., 1863-1864, 150. — Kentropsis hrevi- 
cuspis F.v.M., Ic.—Threlkeldia hrevicuspis F.v.M., B.FL, v., 1870, 198.— Bassia 
hrevicuspis F.v.M., First Census, 1882 and Icon. Austr. SalsoL, Plate Ixvii. 

While working on B. echinopsila F.v.M. my attention was drawn to Echinop- 
silon anisacanthoides F.v.M., which was given as a synonym both by Mueller 
(Fragm.. vii., 14) and Bentham (B.FL, v., 198). I was puzzled by the original 
description, which would not correspond with that of B. echinopsila F.v.M., more 
particularly in the shape of the fruiting perianth and the position of the seed, 
and I came to the conclusion that Mueller had described mixed material. On this 
ground, I decided to reject the name anisacanthoides, which had prior claims, and 
keep to the established name of echinopsila. 

To verify this I had the type specimen of Echinopsilon anisacanthoides sent 
from Melbourne, but on examining this found it to be a form of B. hrevicuspis 
F.v.M. The fruiting perianth of the latter species has a large hollow base, but 
at times the walls of this collapse or wither, or the cavity j^s greatly suppressed. 
This peculiar character of the fruiting perianth is, at times, found on the one 
specimen in company with the normal or typical perianth, so the distinction is not 
even a varietal one, depending probably on the degree of maturity and the drying, 
crueller has figured fine instance of it in his plate of B. hrevicuspis (Icon. Austr. 
SalsoL, PL Ixvii., Fig. 7, perianth on extreme left). The type specimen of 
Echinopsilon anisacanthoides F.v.M. shows this form of fruiting ])eriantli and 
tliercfore is synonymous witli B. hrevicuspis. 



L!Y R. H. AXDJiRSUN. 331 

As B. hrevicnspis F.v.M. and B. echiaopsila F.v.M. closely resemble each 
other in general characters other than the position of tlie seed and the large 
hollow base of the fruiting perianth, it is easy to understand how the original 
confusion arose. 

Seeing- that the species Echinopsilon anisacantJwides F.v.M. was described in 
1858, and Amsa.cantha brevicuspis was not described by Mueller until 1863, and 
as the type specimen and original description are quite plain, we must take up 
the older name of anisacantJioides. I therefore propose calling the species B. 
amsaccmtkoides (F.v.M.), n. comb. 

Bentham preferred to place the species in the genus Threlkeldia, but the 
spines and hardened perianth are quite those of a Bassia. 

Notes in addition to the description. — A small gTeen succulent plant, glabrous 
except for the flowers. Fruiting perianth 5-spined, one of which is bifid, spines 
1-3 mm. long. Leaves linear, rarely more than 12 mm. long. Cavity of the 
fniitiiig perianth very lai'ge. Seed horizontal. Tube smooth and constricted 
near the top. 

Affinity. — In general superficial appearance this species comes close to B. 
echinopsila, with which it is often confused, but the fruiting perianth is very 
different on close examination {see above). It can easily be distinguished from 
other species with large basal cavities by the number of spines and the shape of 
the tube. 

Range. — The species is widely spread in Queensland and to a lesser extent in 
New South Wales. In the National Herbarium are specimens from the follow- 
ing localities: — Mt. Sturt Station, Broken Hill district, Dalby, Q'ld., Boulia, Q'ld. 
I have also seen specimens in the Queensland Herbarium fi'om Roma, Dalby, 
Tower Hill and "Walluinbilla. 

9. Bassia Tatei F.v.M. 

Vict. Nat., vii., 1890, 66; Icon. Austr. Salsol., Plate Ixxi. 

Notes additional to the description.' — An erect perennial, the whole plant 
tomentose, the vestiture on the leaves being longer than on the branches. Finiit- 
ing perianth only moderately hardened, with a large swollen base, the tube marked 
by a few to many longitudinal libs or furrows. Spines weak, divaricate and very 
irregular, 1-3 in number, the longest spine not more than 4 mm. Often with two 
unequal spines, the larger of 'ivhieh is bifid, but at times only one spine is present. 
The limb is fairly well developed. Leaves broad lanceolate or lanceolate, up to 
23 mm. long in the specimens seen. 

Affinities. — The nearest affinities are B. diacantha F.v.M. and B. paralleli^ 
cuspis, n.sp., from both of which it can be distinguished by the soft, swollen, 
ribbed perianth, the unequal iiTegular spines, and the broader leaves. 

Bange. — So far we have only specimens from South' Australia. The speci- 
mens in the National Herbarium are from Mt. Lyndhurst and one from the Ade- 
laide University (locality not specified). 

10. Bassia PxiRALLELicusPis, n.sp. (Plate xxxiv., H.-L.) 

Fruticulus ramosus, ramis tomentosis nonnunquam glabrescentibus, foliis hir- 
sutis lineari-clavatis sessilibus 10-25 mm. longis, floribus solitarlis, perianthii fruc- 
tiferi tube tomentose 3-4 mm. longo ad apicem, 4 mm. lato ad basin oblongam, 
valde excavato cum limbo erecto tomentoso vel piloso 1-2 mm. longo, spinis 2 
parallelibus rigidis gracilibus 2-3 mm. longis usque ad medium pilosis, tertia spina 
minuta vel tuberculo, semine oblique verticali. 

Mt. Lyndhurst, S. Australia, 12/1897 and 8/1899. 



332 lIKVISiOK OF AUSTRALIAN SPECIES OF BASSIA_, 

A small uuder-shrub, prociuubent or sligLtly ascending, the whole plant 
tomentose or hirsute. The vestitui'e on the older branches weai's oft: at times. 
Leaves linear, sessile, the younger ones densely hirsute, the older ones sj)aringiy 
so, up to 25 mm. long by- about 2 mm. wide. Fruiting perianth hardened, tomen- 
tose, with a hollow oblong or oval base, the long axis of which is directed up the 
stem. Spines 2, generally parallel to each other and directed forwards, slender, 
and with a tubercle at the base of one, under 4 mm. long. Tube densely tomen- 
tose, 3-4 mm. long. Limb short, erect, pilose. Seed vertical or oblique. 

Affinity. — This species is allied to B. Tatei F.v.M". from which it can be dis- 
tinguished by the much broader leaves of the former and by the nature of the 
fiiiiting perianth. In B. Tatei F.v.M. the fruiting perianth is softer, larger, 
more irregular and the tube is much ribbed or furrowed and is only slightly 
tomentose, gradually becoming glabi'ous. The spines are divaricate and very 
irregular in number and size, whereas in B. paralleliciispis the spines are parallel, 
more or less equal, and regTilar. 

The species also approaches B. diacantha F.v.M., differing in the position of 
the seed, the base of the fruiting perianth, the nature of the tube, and the fine 
parallel spines. 

Range. — New South Wales, South Australia and Northern Territory. Besides 
the locality already quoted we have specimens from the Finke River between 
CroAvn Pt. and Horseshoe Bend and Tibooburra. The latter specimen is a 
smaller form than the typical one and somewhat more tomentose. 

11. Bassia recurvicuspis W. V. Fitzgerald. 
Mueller Bot. Soc, May, 1904, p. 32.— 2^. Utorali^ Dk'h, Eng. Bot. Jaiirb., 



v::vv.. 1905, 186. Jll 

A glabrous under-shrub, the branches striate. Leaves usually subterete, up 
to 12 mm. long. Tube cylindrical, hardened, hollowed at the apex, up to 4 mm. 
long b}' 2 mm. wide. Spines 4, recurved, two being 2-4 mm. long, the other two 
smaller; the longer spines opposite and diverging, the two smaller approximating 
eaeh other at the base. Very occasionally 5-spined. 

Synonym. — B. litoralis Diels. 

Although I have seen no specimens of Diels's species I am including it as a 
synonym owing to the close similarity between the descriptions of the two species. 
Diels has also furnished a figure v.'hicli agrees with the type specimen of B. 
recurvicuspis W. V. Fitz. The shape of the perianth and the character of the 
spines are apparently identical in both species. 

Affinities. — Its nearest affinity is '»vith B. ecliinopsila F.v.M., from which it 
differs chiefly in the 4 recurved spines and the shape and base of the tube. It 
also resembles B. astrocarpa F.v.M. in the recurved spines, but differs in the 
hardened perianth, glabrousness and number of spines. 

Bmige. — Apparently confined to Western Australia. The type specinaens 
collected ];y W. Y. Fitzgerald (Nannine, 9/1903) are represented m the National 
Herbarium. The type of B. litoralis Diels comes from the vieiiiity of Shark 
Bay. 

12. Bassia glabra F.v.M. 

Kentropsis yluhra F.v.M.. Fragm., i., 139, — AnisacanUia ijluJtra F.v.M., B.Fl., 
v., 1870, 200.— B. glabra, F.v.M., First Census, 1882; Icon. Ausiv. Salsol., 
Plate Ixvi. 

Notes addilional to the dcstcriptifni. - A ghi]ji'<)u> under-sliiul) witli linear 



BY R. H. ANDERSON. 333 

leaves up to 11 mm. long in the specimens seen. Fruiting- perianth small, about 
2 mm. long, ribbed vertically, with 2 equal, opposite, diverging spines (often 
almost horizontal), usually with a stout tubercle or third small spine present. 
Limb erect and hardened, forming a distinct ridge to the top of the perianth. 
The spines in the only specimen I have seen (Sturt's Creek, the type) are 1-2 
mm. long. The seed is vertical. 

I cannot altogether follow Mueller's fignre (Icon. Austr. SalsoL, Plate Ixvi.). 
In Fig. 7 he has a section of the perianth which appears to be closer to that of 
B. uniflora (R.Br.) F.v.M. and which shows a horizontal seed. Although Mueller 
placed this species in the genus Kentropsis which has a horizontal seed, Bentham 
gives the position of the seed as vertical or oblique, and the seed bif the type is 
certainly vertical. I have not seen anything approaching Figs. 1 and 6 which 
show the fruiting perianth 'with very long spines, somewhat resembling those of 
B. longicuspis F.v.M. 

Mueller himself gives the length of the spines (Fragm., i., 139) as 1-2 lines 
long, and in the specimen examined I found them not to exceed 2 mm. 

The species, however, is very scantily represented and additional material is 
necessary in order to define the species thoroughly. 

Affinity. — Closely resembles B. echinopsila F.v.M. in regard to the foliage, 
but the fruiting perianth differs in the number of spines, in general shape, and 
in the base. B. echinopsila F.v.M. is 5 — 6-spined. 

Range. — I have seen only one specimen (Sturt's Creek, F.v.M., the type). 

13. Bassia bicuspis F.v.M. 

Anisacantha hicuspis F.v.M., Trans. Viet. Instit., 1855, 133; Hook. Kew 
Journ., viii., 204; B.Fl., v., 1870, 200.— 5. bicuspis, F.v.M., First Census, 1882; 
F.v.M., Icon. Austr. SalsoL, Plate Ixxiii. 

Notes in addition to the description. — Plant quite glabrous, except for the 
limb which is shortly tomentose or pubescent and a ring of hairs at the base of 
the perianth. Fruiting perianth with 2 spines up to 11 mm. long and a tubercle 
or small spine about 1-1-| mm. long. Limb up to 5 mm. long, much exceeding 
the length of the tube which is usually 14-2-| mm. Brandies much scarred by 
the falling off of the fruiting perianths and leaves. 

This species is readily distinguishable from its allies by the glabrous leaves, 
the much greater relative leng-th of the limb to the tube, and the spines. 

Range. — The only specimen I have seen is the type (Salt Plains, Cudnaka, 
F.v.M.). The other specimen mentioned by Bentham (B.FL, v., 200) from be- 
tween the Stokes Range and Cooper's Creek (Wheeler) is probably B. longicuspis 
F.v.M. 

14. Bassia longicuspis F.v.M. 

Viet. Nat., ix., 1893, 187. 

Notes in addition to the description. — Glabrous except for the flowers, limb, 
and axils of the leaves which are more or less hirsute. Leaves succulent, often 
2-3 cm. long, and sometimes exceeding 4 cm., 3-4 mm. in width. Spines of the 
fruiting perianth usually 3-4 in number, very unequal, the longest at times ex- 
ceeding 35 mm. ; 1 or 2 spines generally much shorter than the others. Seed 
vertical. Stems and spines often reddish in colour. 

Synonyms. — The type specimen (Charlotte Waters, Rev. Kempe, 1885) is 
labelled B. hicuspis var. longicuspis and underneath B. longicuspis. Bentham 
(B.FL, v., 200) quotes a specimen from between Stokes Range and Cooper's 



334 REVISION OF AUSTRALIAN SPECIES OF BASSIA, 

Creek (Wheeler) under Anisacantha hicuspis F.v.M., and gives the perianth as 
"fully 3 lines long and the longest spine |-1 inch." 

This specimen quite probably belongs to B. longicuspis F.v.M. MueUer 
(Fragm., vii., 14) gives a var. longicuspis of Anisacantha erinacea with spines 
almost an inch long. The locality is given as Stokes Range, so the probability 
is that it is the same specimen as quoted by Bentham under B. hicuspis. 

Affinity.— K\i\wag]i originally placed under B. hicuspis F.v.M., it differs 
greatty from that species in the spines, shape of the tube, and the relative length 
of the limb and tube. Its affinity rather lies with B. tricuspis (F.v.M.), n. comb., 
from which it can be distinguished by the 4 unequal spines, the scarcely dis- 
tended base of the tube, and the length of the leaves and spines. 

I have seen a specimen from the Queensland Herbarium, collected by Miss 
M. J. Bancroft at LongTeaeh, which appears to be a weak, less robust form of 
B. longicuspis F.v.M. and intermediate between that species and B. tricuspis 
(F.v.M.). 

Range. — The following localities are represented in the National Herbarium :— ^ 
S. Australia: Moolooloo Station between Beltana and Blinman, Mt. Lyndhurst, 
Charlotte Waters, the type; N.S.W.: Broken Hill District, White Cliffs, and 
Barrier Ranges. 

15. Bassia divaricata (R.Br.) F.v.M. 

Anisacantha ddvaiicata R.Br., Prodr., 1810, 410. — Anisacantha erinacea Moq., 
B.C. Prodr., xiii., ii., 1849, 122. — Chenolea tricuspis F.v.M,, Fragm., x., 92 
(partljO? but not Anisacantha tricuspis F.v.M., Trans. Vict. Instit., 1855, 133. — 
B. divaricata, F.v.M., First Census, 1882. 

In the material labelled Bassia divaricata in the National Herbarium I found 
there Avere evidently 2 distinct species which for convenience may be termed A. 
and B. 

The two can be separated by: — 

(1). the shape and method of attachment of the fruiting perianth. In 
species A. the base is very oblique so that the perianth appears to be attached by 
its side, the tube being roughly parallel to the stem. In species B. the base of 
the fruiting perianth is broad and swollen, the tube roughly at right angles to 
the stem. 

(2). the limb, which is erect in B. and curved over in A. 

(3). the number and nature of the spines. A. has 3 unequal horizontally 
spreading or slightly recurved spines with usually a fourth short spine at the 
base of one of the others. B. has 3 somewhat ascending spines, more or less 
equal, without the fourth short spine of species A. Very occasionally a fourth 
spine is present, but it is then equal to the other 3 spines, and all placed at 
equal distane&s from one another. 

(4). the position of the seed which in A. is always vertical, whereas in B. 
it is usually oblique or obliquely vertical or at times approaching the horizontal. 

Species A. is evidently Anisacantha divaricata R.Br, as in the original 
description (Prodr., 410) Brown describes the perianth as with 4 unequal spines. 
I am therefore accepting species A. as the true B. divaricata (R.Br.) F.v.M. 

Species B. is identical with the type specimen of Anisacantha tricuspis 
F.Y.M. (Trans. Vict. Instit., 1855, 133) from the Murray River. 

Anisacantha erinacea Moq.is evidently a synonym of B. divaricata (R.Br.) 
F.v.M., as the species is described (D.C. Prodr., 13, ii., 122) with 3 spines, one 
of which is at times bifid. This would apply to B. divaricata (R.Br.) F.v.M. 



BY R. H. ANDERSON. 335 

Moquin was evidently somewhat confused in regard to Anisacantha divaricata 
K.Br., as he describes it (D.C. Prodr., 13, ii., 122, immediately preceding A. 
erinaeea) as with villose leaves. Although I have not seen the type of A. erinacea 
Moq. it seems safe to assume that it is synonymous with B. divaricata (R.Br.) 

r.v.M. 

This leaves the way clear to take up Mueller's name of trieuspis for the 
species B., calling it B. trieuspis (F.v.M.), n.eomb. 

The differences between the two species are apparently constant, and they 
can readily be separated. Mueller (Icon. Austr. Salsol., Plate Ixxvii.) has figured 
B. trieuspis (F.v.M.) together with the 4-spined form. 

Notes on B. divaricata (R.Br.) F.v.M. — Grlabrous except for the flowers, 
axils of the leaves, and the limb. Longest spine usually about 8-15 mm. long, 
but occasionally up to 20 mm. Viewing the fruiting perianth from above, very 
often 1 or 2 of the spines appear to be twisted in an anti-clockwise direction. 

Range. — A not uncommon species in the drier parts of Western Australia. 
South Australia, ISTew South Wales, and Queensland. The following localities are 
represented in the National Herbarium : — W. Aust. : Grwalia, Cue ; S. Aust. : Mt. 
Lyndhurst, Quorn, Woolsbed Flat; N.S. Wales: Broken Hill, Paroo R., Barrier 
Ranges, Tibooburra, Wilcannia, Warrego R., Brewarrina Downs. 

16. Bassia tricuspis (F.v.M.), n.eomb. 

Anisacantha tricuspis F.v.M., Trans. Viet. Instit., 1855, 133, and in Hook. 
Kew Joum., viii., 204. — Chenolea trieuspis F.v.M., Fragm., x., 92 (partly). 

I have already detailed the distinctions between this species and B. divaricata 
(R.Br.) F.v.M. under the latter species. The erect limb, the 3 more or less 
equal spines, and the swollen base of the fruiting perianth, with the tube at right 
angles to or inclined to the stem are the chief distinguishing features of the 
species. 

The branches are glabrous and striate. The seed is usually obliquely vertical 
but I have seen some specimens in which it is almost horizontal. 

Range. — I have seen specimens from Queensland, South Australia and New 
South Wales, the following localities being represented in the National Her- 
barium: — New South Wales: Hay, Paroo R. District, Warrego R., (mixed with 
B. divaricata), Darling Downs (Lau?, labelled A. erinacea by Mueller); Queens- 
land: Tower Hill. 

The specimen from the Suttor R. (E. BoVman) mentioned in B.FL, v., 200 
under Anisacantha divaricata R.Br, belongs to B. tricuspis (F.v.M.). It is 
labelled A. erinaoea. 

17. Bassia lanicuspis F.v.M. 

Anisacantha lanicuspds F.v.M., Fragm., ii., 170. — B. lanicuspis, F.v.M., 
First Census, 1882. 

I have already detailed the nature of this species and its distinction from 
B. eriacantha (F.v.M.), n.eomb. when dealing with the latter species. 

The fruiting perianth is enveloped in long silky hairs; the tube is oblong 
or turbinate; spines 2-4, slender, weak, hirsute, and up to 8 mm. long or some- 
times longer. The leaves have a vestiture of long silky hairs. Branches tomen- 
tose or becoming almost glabrous. Leaves usually under 15 mm. long but oc- 
casionally longer. 

Range. — Occurs in Queensland, New South Wales and South Australia. The 



336 revisiojSt of Australian species of bassia^ 

following localities are represented in the National Herbarium: — Queensland: 
Georgina R. ; New South Wales: White Cliffs, Barrier Ranges, Brewarrina, 
Warrego R., Coolabah, Silverton; South Australia: Mt. Lyndhurst, Beltana. 

18. Bassia Drummondii (Benth.) F.v.M. 

Anisacantha Drummondii Benth., B.FL, v., 1870, 199. — B. Drummondii, 
First Census, 1882. 

Under this species I include Anisacantha hispida Spencer Moore (Journ. 
Linn. Soc. Lon., xlv., 1920, 190) as a variety, the only difference between the 
two species being that one is more hispid than the other. In all other respects, 
both in the material and the descriptions, the two species are identical. Both 
are marked by the characteristic scarring of the branches by the hardened base 
of the leaves, and the fruiting perianths and spines are the same. The leaves 
and habit of the two species are also identical. 

The type specimen of Anisacantha hispida S. Moore is in the National Her- 
barium but no type of Bassia Drummondii (Benth.) F.v.M. The Melbourne 
Herbarium was unable to furnish the latter, but sent a specimen of B. Drum- 
mondii, determined by Mueller, which proved to be identical in all respects with 
the type of A. hispida. 

Bentham described Anisacantlia Drummondii as with "a few long spreading 
hairs on the leaves." 

The National Herbarium contains a specimen (Kalgoorlie, 8/1898, W. V. 
Fitzgerald) which is glabrous except for a few hairs in the axils of the leaves, 
flowers and limb. 

In the hispid form the older leaves are very often almost or quite glabrous, 
the younger ones towards the ends of the branches being densely invested with 
long hairs. But the various forms are obviously the same species, lesembling 
each other in the several points which characterise the species and separate it 
from its congeners. There is thus a species which varies from an almost 
glabrous form to a densely hispid form, but otherwise maintains the features 
which distinguish it specifically. A parallel case is that of B. quinquecuspis 
F.v.M. 

We are not therefore justified in distinguishing the forms specifically and 
I propose to reduce Anisacantha hispida S. Moore to Bassia Drummondii (Benth.) 
F.v.M., var. hispida. 

Descriptive notes on the species. — B. Drummondii may be distinguished from 
other members of the genus by the small obtuse leaves which, on falling off, 
make prominent scars on the branches. In this respect it resembles B. hicusj^is 
F.v.M. but differs from that species in the nature of the fruiting perianth and 
spines. 

The fruiting perianth is very small, the tube 1-2 mm. long, and furnished 
with 3-4 spines, two of which are long and spreading, and up to 9 mm. in 
length, the remaining ones much shorter or one reduced to a tul)ercle. The seed 
is usually vertical, but occasionally it is obliquely horizontal. 
Var. HISPIDA. 

Distinguished from the species by the densely hispid leaves, more particularly 
towards the ends of the branches, but merges into the almost glabrous form of 
the species. 

Bange. — The .species and its variety are apparently confined to Western 
Australia. The following localities are represented in the National Herbarium: — 



BY R. H. ANDERSON. 337 

Kalgoorlie (A mixture of the var. hispida and the almost glabrous form), MuUine 
(the type of A. hispida S. Moore), Kalgoorlie (var. hispida), 12 mis. E. of 
Kaaowna (var. hispida) and bel^ween the Tipper Blackwood R. and Lake Lefroy 
(var. hispida) . 

19. Bassia decurrens J. M. Black. 

Trans. Roy. Soc. S. Aust., xlvi., 1922, 567. 

A species distinguished from its allies by the pilose leaves, and the ribbed 
perianth tube bearing 2 divergent spines, one of which has, at or close to its 
base, 1-3 short spines or tubercles, which are also associated with a prominent 
rib of the tube. In the specimens I have examined, the tubercles or small spines 
are generally 1 or 2 in number and only occasionally 3. The longer spines 
sometimes up to 11 mm. long. Its nearest affinity is B. hieuspis F.v.M., which 
it resembles in the short tube and the well developed limb, but differs in the 
pilose leaves, the tubercles or short spines at the base of the larger spine and the 
prominent rib to the tube. B. hieuspis has glabrous leaves, only one tubercle, 
and the tube is scarcelj?^ ribbed. 

Range. — So far this species is only known from South Australia and New* 
South Wales. The following localities are represented in the National Her- 
barium: — Port Augusta, S.A. (the type). Broken HiU District, Paroo River 
District. 

20. Bassia obliquicuspis, n.sp. (Plate xxxiv., D.-Gr.) 

Sderolaena diacantha var. longispina Benth., B.FL, v., 1870, 195 (in part). 

Fruticulus ramosus, ramis tomentosis vel lanatis, foliis lanatis lineari-clavatis 
sessilibus 8-15 mm. Icngis, floribus solitariis pilosis, styHs duobus, perianthii fructi- 
feri tubo tomentoso subcylindrico 2-4 rnm. longo, cum base oblonga, leviter excavato, 
limbo brevissimo, spinis 2 supra medium tomentosis divergentibus obHque curvatis 
plus minusve aequalibus 4-8 mm. longis quarum una in tubsrculum decurrit, 
semine verticali. 

Found in the western parfs of New South Wales and in South Australia. 
N.S.W.: Barrier Ranges, Broken Hill district, Cobar; South Australia: Black 
Bluff, Murray Flats, Poii Augusta. 

An under-shrub, the whole plant densely tomentose or lanate. Leaves linear 
clavate, somewhat thick, up to 15 mm. long but usually shorter, and about 1-2 
mm. in 'width. The fruiting perianth has an oblong furrowed base, spreading at 
right angles to the tube along the stem and often equal in length to the tube. 
This spreading base is 2-4 mm. long by li-2f mm. wide. The tube is up to 4 
mm. long. The 2 divergent spines are tomentose for about two-thirds their 
length, and are curved and always twisted obliquely away from each other, that 
is, they are not in the same vertical plane, but point in opposite directions from 
it. The base of one spine is always marked by a short thick tubercle. 

In the past this species has been mixed up with B. diacantha F.v.M., but 
the two species have few points of resemblance. More particularly, the nature 
of the tube and the position of the seed are quite different in the two species. 
B. diacantha has a much hollowed base to the tube, and the seed is horizontal or 
oblique. B. ohliquicuspis has only a furrow at the base of the tube, and the s^ed 
is always vertical. The ribbing and general shape of the tubes of the two 
species are also quite different. The oblong base extending at right angles to the 
tube characteristic of B. ohliquicuspis is never found in B. diacantha. 

Bentham (B.FL, v., 195) makes a var. longispina of B. diacantha. but queries 
it. I have seen one of the specimens quoted by him, namely, Gawler's Range 



338 REVISION OF AUSTRALIAN SPECIES OF BASSIA^ 

(Sullivan), and this is a mixture of B. ohliquicuspis and the closely allied species 
B. patenticuspis, n.sp. In his description of the variety he quotes the perianth 
as very tomentose or almost glabrous. The very tomentose perianth refers to 
B. ohliquicus2jis and the nearly glabrous perianth to B. patenticwspis. 

I have refrained from taking up the name longispina, as both the descrip- 
tion and the material are mixed. In addition, a similar name is already oc- 
cupied by B. longicuspis F.v.M. 

The fact that both material and description are mixed affords one legitimate 
grounds for not adopting the prior varietal name, which is not compulsory. 

Prof. A. J. Ewart (Proe. Roj^ Soc. Viet., xxvi., i., 1914) placed the var. 
longispina under B. lanicuspis F.v.M., pointing out that no special varietal 
distinction was now necessary. The specimens, however, have little resemblance 
to the two species hitherto grouped under B. lanicuspis^ differing from B. eria- 
cantha (F.v.M.), n. comb, in the holloi\v base of the tube and the horizontal seed 
of that species, and from B. lanicuspis proper by the number of spines and the 
long silky hairs about the flowers and fruiting perianth of the latter species. 

In addition to the localities already quoted for B. ohliquicuspis there is a 
dwarf form with the tube 1-3 mm. long and the spines not exceeding 3 mm., 
and the leaves short and densely clustered. (Broken Hill, A. Morris). 

21. Bassia PATENTICUSPIS, n.sp. (Plate xxxiv., A.-C.) 

Sclerolaena diacantha var. longispina Benth., B.FL, v., 195, in part. 

Fruticulus ramosus, ramis tomentosis nonnunquam glabrescentibus, foliis pilosis 
lineari-clavatis sessiHbus 6-10 mm. longis, floribus solitariis, perianthii fructiferi 
tubo subcylindrico parce tomentoso circ. 2 mm. longo cum base ovata vel oblonga, 
limbo erecto 1 mm. longo, spinis 2 divergentibus non oblique curvatis 3-7 mm. 
longis quarum una in tuberculum decurrit, semine verticali. 

New South Wales : Broken Hill, Barrier Ranges ; South Australia : Baroota, 
Port Augusta, Beltana; Wiestem Australia: Nannine. 

An under-shrub closely approaching B. ohliquicuspis, n.sp., but readily 
separable from that species. The chief points of difference lie in: — 

(1). The spines which in B. ohliquicuspis are always twisted away from 
each other, apart from being divergent, so that they are not in the same vertical 
plane. In B. patenticuspis they are not twisted, but are both in the same vertical 
plane. This feature is constant in all the specimens examined by me. 

(2). The limb which in B. ohliquicuspis is very short, so as to be hardly 
noticeable, whereas in B. patenticuspis it is erect and well developed and ap- 
proximately half the length of the tube. 

(3). The shorter tube and less furrowed smaller base of B. patenticuspis. 

(4). The vestiture. The stems of B. patenticuspis are not so densely tomen- 
tose and the fruiting perianth is markedly less tomentose than those of B. 
ohliquicuspis. At times the tube of the former species is almost glabrous. 

(5). Habit of growth. Although I have had no opportunity of studying 
the two species in the field myself. Miss Collins has told me that, when examining 
the plants in the Broken Hill district, she could readily separate them on their 
growth habit. 

At first I intended making one a variety of the other but the differences are 
so constant that a specific name seems justified. 

22. Bassia echinopsila F.v.M. 
Anisacantha echinopsila F.v.M., Fragm., vii., 14 and B.FL, v., 201. — Cheno- 
lea echinopsila, Fragm., x., 92. — B. echinopsila, First Census, 1882; Fragm., xii., 
12. 



BY R. H. ANDERSON. 339 

This species has been more or less confused with B. anisacanthoides (F.v.M.), 
n. comb, and B. convexula, n.sp. Although both Mueller and Bentham placed 
Echinopsilon anisacanthoides F.v.M. as a synonym of Anisacantha echinopsila 
F.V.M., an examination of the type showed clearly that it is a form of A. hrevi- 
cuspis F.v.M. 

Notes in addition to the description in B.Fl., v., 201. — An extremely variable 
plant in regard to its vestiture. We have a perfectly glabrous form, a form 
with the young shoots hirsute or pilose, and one in which all the leaves are 
densely covered with long hairs. Sometimes only the young leaves at the tips 
of the branches are slightly hirsute. The various forms merge so gradually into 
one another that it seems almost impossible, and of little advantage, to dis- 
tinguish the hirsute form as a iwell-defined variety. 

The tube of the fruiting perianth is oblong or cylindrical, up to 3 mm. 
long, ribbed or furrowed longitudinally and with a grooved base with two short 
spurs. 

The spines are usually 5, with one spine bifid, varying from 1-5 mm. long, 
but occasionally there are 6 separate spines. They vary from almost erect to 
horizonta!lly spreading, but are usually inclined upwards. The seed is truly 
vertical. 

Affinity.- — In general appearance this species bears a remarkable resemblance 
to B. anisacanthoides (F.v.M.), n. comb., and often can only be distinguished by 
a close examination of the fruiting perianth. In B. anisacanthoides the tube is 
hollow and the seed placed horizontally. In B. echinopsila the tube is not hollow 
and the seed is placed vertically. The shape and ribbing of the tube are also 
different. 

Mudler's figure (Icon. Aust. Salsol., PL, Ixix.) is a very mixed one. The 
extreme left hand figure of the series of fruiting perianths, marked 6, is that of 
B. anisacanthoides (F.v.M.), n. comb. ; Fig. 2 also appears to be the latter species. 
The extreme right hand one of fig. 6 is B. convexula, n.sp., as are also the top 
and lower left hand figures of series 7. 

Range.- — 'This species has a fairly wide range in Ndw South Wales and 
Queensland. The following localities are represented in the National Herbarium — 
Queensland: Longreaeh, Blackwater E. of Emerald, Rockhampton (O'Shanessy) 
and Crocodile Creek (BoAvman). The last two specimens are quoted by Bentham 
(B.FL, v., 201); Neiv South Wales: Nyngan, Brewarrina, Namoi R., Cutbaroo 
Siding, Tibooburra, Cobar, Warrego R., Burren Junction, Mooculta Siding, 
Botanic Gardens, Sydney (introd.). 

23. Bassia tubata^ n.sp. (Plate xxxv., A.-C.) 

Fruticulus, ramis erectis gracilibus tomentosis vel glabrescentibus, foHis lineari- 
subteretibus sessilibus 5-15 mm. longis, floribus solitariis, stylis 2-3, perianthii fruc- 
tiferi tube cylindrico parce tomentoso 3-4 mmi. longo cum base vix dilatata ex- 
cavataque, limbo erecto brevissimo, spinis 4 quarum una bifida erecta, ceteris hori- 
zontale divergentibus, 3-5 mm. longis, semine verticali. 

Coonamble (E. Breakwell, 1918 and R. McDiarmid, Feb., 1922). 
A small under-shrub with numerous slender erect branches, the whole plant 
from fairly densely hirsute to almost glabrous. Leaves linear lanceolate or sub- 
terete, up to 15 mm. long by l-lj mm. wide. Styles usually 3, or occasionally 2. 
Fruiting perianth more or less tomentose with a cylindrical tube, the base of 
which is usually somewhat dilated and hollow. Tube 3-4 mm. long by li-2 mm. 
wide, the base shortly oblique, but the tube free from any lateral attachment to 
the stem. Spines 4, one of which is more or less erect and bifid, the remaining 



340 REVISION or Australian species of bassia^ 

three spreading more or less horizontally, all about 3-5 mm. long except the 
smaller of the bifid spines which is 2-3 mm. Base of the perianth is usually 
somewhat hollowed, but at times this is not noticeable. The seed is vertical. 

The only specimen of this species in the National Herbarium is one col- 
lected by Mr. E. Breakwell at the Coonamble Experiment Farm. I got into 
touch with Mr. McDiarmid, the manager of that Fann, and through him ob- 
tained additional specimens which showed that the specific differences were con- 
stant. He informed me that it is only to be found in places protected from 
stock. 

The distribution of the species is peculiar, being so far found only at Coon- 
amble. One would expect it to be found in similar localities in New South 
Wales. 

The long cylindrical tube, free from lateral attachment to the stem and with 
a slightly hollow base, and the 4 spines, one being bifid and erect, the others 
horizontal, distingTiish it from its congeners with a vertical seed. ' Its nearest 
affinity is the next species, B. intricata, n.sp., from which it differs in the slightly 
dilated and hollow base of the fruiting perianth, the arrangement of the spines, 
the method of attachment of the fruiting perianth, and the vestiture. B. intricata 
is glabrous except for the flowers and axils of the leaves, and the fruiting 
Derianth is attached for the greater part of the length of the tube by the very 
obHque base. 

24. Bassia intricata^ n.sp. (Plate xxxv., D.-F.) 

Fruticulus ramosus, ramis glabris intricatis, foliis lineari-clavatis vel sub- 
teretibus sessilibus 5-15 mm. longis, fioribus solitariis, perianthii fructiferi tube 3-5 
mm. longo sub-cylindrico cum base valde oblique, limbo erecto hirsuto 1-2 mm. 
longo, spinis 5, vel 4 quarum una bifida, 5-12 mm. longis saepe recurvatis, semine 
verticali. Approximat B. divaricatae F.v.M. 

Corona (Miss M. Collins), Mt. Lyndhurst, S.A. (Max Koch). 

An intricately branched compact under-shrub, glabrous except for the hir- 
sute flowers, limb,, and for dense tufts of long hairs in and around the axils of 
the leaves. Leaves linear clavate or semiterete, but sometimes broader, sessile, 
5-15 mm. long, succulent. Flowers solitary, styles 2. Tube of the fruiting 
perianth cylindrical, 3-5 mm. long with a very oblique base. Spines 5, or 4 
Avith one bifid, slender, usually more or less recurved, up to 12 mm. long, but 
generally shorter. One spine usually much shorter than the others. Limb erect, 
hirsute, 1-2 mm. long. Seed vertical. 

This species in the past has been called both B. quinquecuspis F.v.M. and 
B. (Uvaricata (R.Br.) F.v.M., but its affinities lie with the latter species. In 
general habit it ajDproaches B. divaricata very closely, but can be separated from 
that species b}^ the number of spines (5 in B. intricata and 3-4 in B. divaricata), 
the erect and longer limb of B. intricata which in B. divaricata is always re- 
curved or bent over to one side, the denser tufts of hairs in the leaf axils and on 
the flowers, and the more scattered fruits. The tube of the fruiting perianth of 
B. intricata is also generally longer and less tapering than that of B. divaricata. 

While recognising the close resemblance between these two species, the 
differences I have enumerated are constant and appear to be specific rather than 
varietal. 

The long tube and the limb at once distinguish the species from B. quinque- 
cuspis. 



BY R. H. ANDERSON. 341 

Range.— In addition to the two localities already quoted there are specimens 
in the National Herbarium from: — Milparinka, Mutooroo, S.A., Lake Eyre, S.A., 
Mt. Nor. "West. The spines in the last-named specimen are very strongly re- 
curved. 

25. Bassia quinquecuspis F.v.M. 

Anisacantha quinquecuspis F.v.M., Trans. Viet. Inst., 1855, 134 and Hook. 
Kew Journ., viii., 204. — A. muricata Moq., Chenop. Enum., 1840, 84, and in 
D.C. Prodr., xiii. (ii.), 122. — A. gracilicuspis F.v.M., Fragm., ii., 170. — B. 
quinquecuspis, First Census, 1882; Icon. Aust. Salsol., PL Ixxvi. 

This species is one of the most variable within the genus. In the typical 
form the plant is glabrous and intricately branched, the branches usually striate; 
leaves linear-lanceolate, acute, glaucous, and up to 25 mm. long, but usually 
about 15 mm. 

The tube is very shortly cylindrical, 1-2 mm. long, and with a very oblique 
base firmly attached to the stem and only broken away with difficulty. Spines 
5, or 4 with one spine bifid, unequal, the longest up to 15 mm. 

In the var. villosa the plant is more or less hirsute or tomentose, the vesti- 
ture on the leaves consisting of long hairs, and. the branches sparingly tomentose. 
The fruiting perianth is also more or less tomentose. The leaves are usually 
broader and shorter than those of the glabrous form of the species and the spines 
shorter. But the differences appear to be only varietal. Some specimens are 
only sparingly hirsute and gradually lose their vestiture, merging into the 
glabrous form. 

The position of the seed of B. quinquecuspis is also a variable character. It 
may be truly vertical, oblique, or almost horizontal, but the oblique condition is ' 
most common. Generally speaking the obliquely horizontal position is most 
common in the var. villosa, and the vertical position in the glabrous form. But 
this is not a hard and fast rule, and so cannot be taken to separate the forms 
specifically. The peculiar irregular character of the fruiting perianth probably 
accounts for the variation. 

I have seen the type specimen of A. gracilicuspis F.v.M. and it agrees with 
var. villosa. Bentham placed A. gracilicuspis as a synonym of var. villosa and, 
although the name gracilicuspis was the first established, we must retain Bent- 
ham's name villosa for the variety. 

Vernacular name of B. quinquecuspis. — This species is widely known as 
Roley-poley. The older plants become detached from the ground and are driven 
along \>j the wind, gradually forming a big ball by the addition of other plants. 
This is finally brought up against some obstacle, chiefly fences and trees, where 
it piles up, forming a thick mat on decomposition. 

Affinity. — This species approaches B. cMvaricata (R.Br.) F.v.M., differing 
in the number and nature of the spines and in the shorter tube. The var. villosa 
approximates closely to B. Birchii F.v.M., under which species the relationship is 
discussed. 

Range. — Probably the most common species of Bassia in New South Wales. 
It is also found in Queensland and Victoria. In the National Herbarium the 
typical form is represented from the following localities : — Coally nr. Milparinka, 
Tibooburra, Burren Junction, Gunnedah, Moree, Denman, Warrego R., Bangate, 
Berrawinia Downs, Junction bet. Murray and Darling R., Mt. Harris, Laidley. 

Var. villosa Benth. is represented by the following localities: — Bathurst, 



342 REVISION OP AUSTRALIAN SPECIES OF BASSIA^ 

Outbaroo Siding, Forbes, Cobar, Condobolin Fla,ts, Wanganella, Blacktown, 
Flemington, and Camden Municipality. 

26. Bassia Forrestiana F.v.M. 

Chenolea Forrestiana F.v.M. coll. — B. Forrestiana, Fragm., xii., 1882, 12; 
Icon. Austr. Salsol., PI. Ixxv. 

Notes in addition to the description. — A fairly robust under-shrub with more 
or less open branches, the whole plant tomentose or hirsute. Leaves linear, re- 
curving towards the apex and broadening slightly at the base, long, at times up 
to 30 mm., and 2-2J mm. in width. Tube of the fruiting perianth sub-globose 
or depressed, about 2-3 mm. long, the limb erect and about half the leng-th of 
the tube. The fruiting perianth has 2 long spines, glabrous exccipt towards the 
base, and up to 25 mm. long, and 1-2 or very rarely 3 much shorter spines under 
7 mm. long. The seed is hori^sontal or oblique. 

I have seen only the type collected by J. Forrest at the Grascoyne River, 
Western Australia, in 1882. Mueller's plate (Icon. Austr. Salsol.) gives an ex- 
cellent representation of the species. 

The species can be separated from its allies by the very short tube, the long 
spines, the dense vestiture and the somewhat thickened, apically recurved leaves. 

Its nearest affinity is probably B. limbata J. M. Black, which it resembles in 
the character of the leaves and the short tube. It differs from this species, how- 
ever, in the much longer, more slender, spines and in the presence of the 1-3 
shorter spines. B. limhata has only one additional very small spine and the tube 
is stouter. 

27. Bassia ventricosa J. M. Black. 

Trans. Roy. Soc. S.A., xlvi., 1922, 566. 

A more or less hirsute under-shrub. The tube is almost globular or sub- 
cylindrical, hirsute, and bearing two opposite divergent spines and one or two 
shorter spines. In the majority of our specimens there is only one shorter 
spine, but two are not uncommon, and very occasionally a third one is present. 
The seed is horizontal or obliquely so. 

It is quite a distinct species, being readily separable from its allies by the 
shape of the tube and the nature of the spines. 

Bange. — The species enjoys a fairly wide range in the drier parts of New 
South Wales and South Australia. The following localities are represented in 
the National Herbarium: — 

N.S.W.: BaiTier Ranges, Campbell's Cr. Broken HiU district, Tibooburra; 
S. Aust.: Port Augusta (J. M. Black 9/1920. The type) and Mt. Lyndhurst. 

28. Bassia limbata J. M. Black. 

Trans. Roy. Soc. S.A., xlvi., 1922, 567. 

This species is allied to B. Forrestiana F.v.M. and B. hicornis (Lindl.) 
F.v.M. It seems to merg© to some extent into the latter species but differs 
chiefly in the smaller, not so thickened perianth tube, the much more con- 
spicuous limb which is often equal in length to the tube, and in the leaves being 
always tomentose or hirsute. In B. bieornis the leaves tend to become glabrous. 
The third spine on the face of the perianth is present in both species. 

It is apparently confined to western New South Wales and South Australia, 



BY R. H. ANDEBSOiv^. 343 

the following localities being represented in the National Herbarium : — Barrier 
Ranges, Mundi Mundi Plains, Broken Hill district, Campbell's Cr., Broken Hill 
district; we have 2 specimens (Tibooburra, 0. Couch and Mt. Ljrndhurst, S.A. 
Max Koch), which appear to be intermediate forms between B. limbata J. M. 
Black and B. hicornis (Lindl.) F.v.M. 

29. Bassia bicorn"is (Lindl.). 

Sclerolaena hicornis Lindl., Mitch. Three Bxped., ii., 1838, 47. — Kentropsis 
lanata Mocj., Chenop. Enum., 83 and in D.C. Prodr., xiii. (ii.), 138. — Anisacantha 
hicornis I'.v.M., Fragm., vii., 14:.— B. hicornis^ First Census, 1882; Fragm., sii., 
1882, 12; Icon. Austr. Salsol., PL Ixxix. 

One of the more commonly known species of Bassia and often termed "Goat- 
bead." It is a very variable plant, the typical form having a woolly white 
tomentum on the branches and fruiting perianth, but the leaves only slightly 
hirsute or eventually becoming glabrous. At times the leaves are very long, be- 
ing up to 40 mm., and 3 mm. in width, but usually they are much less than this. 
The tube is very woody, varying from 3 to almost 6 mm. long, the limb erect 
and about 2 mm. in length. The fruiting pierianth is 2-spined, the spines vary- 
ing from 5 to 20 mm. long, and often a third spine or tubercle is present on 
cne fa,ce of the tube. This third spine at times reaches 5 mm. in length, but 
is usually much reduced. 

A not uncommon form of the species is an apparently stunted variety, in 
■which the spines, fruiting perianths and leaves are all considerably reduced. The 
spines in this form are often only 3-4 mm. in length, the tube is not so much re- 
duced, but the leaves remain more or less densely hirsute. The persistent hairs 
on the leaves can also be found in more typical B. hicornis. 

This stunted form may possibly be made a variety, but it appears to depend 
only on local and seasonal variations. Mr. J. Mitchell, of Walgett, who has 
studied the species in the field, writes that the form is "stunted on account of 
the dry conditions; under ordinary seasonal conditions it is greatly increased in 
the size of the plant, leaf and fruit, the two colors, viz., green and white, being 
much more marked." 

Intermediate forms between the two extreme types are also to be found. 

Range. — This species is apparently fairly common in the drier parts of 
Western Australia, South Australia, New South Wales, Northern Territory, and 
Queensland. The following localities are represented in the National Her- 
barium: — Qld.: Boulia, Longreach; N.S. Wales: Nevertire, Wilcannia, Coolabah, 
Bourke district, Nyngan, Wanaaring to Uriseno, Baneanya; S. Aust.: Lake Eyre; 
W. Aust.: Barwin R. ; N. Terr.: Henbury Station, Finke River. 

The stunted form is represented by the following localities : — Walgett, 
Warrego R., Darling R., North-west N.S.W., Brewarrina, Namoi R. 

30. Bassia Cornishiana F.v.M. 

Austr. Chemist, 1885. 

Notes on the species. — An under-shrub, the whole plant densely woolly to- 
ir.entose or hirsute. The leaves are broadly lanceolate or obovate and up to 15 
mm. or mone long. The fruiting perianth is very broad and flat on top, 3-6 mm. 
across, the tube short, 2-4 mm. long and tapering to a circular or ovate base. 
Spines usually 6, rarely 5, horizontally spreading or slightly recurved, the longest 
rarely more than 6 mm. Commonly 5 of the spines are arranged at equal dis- 



344 REVISION OF AUSTRALIAN SPECIES OF BASSIA^ 

tanoes around the fruiting perianth, and axe roughly equal in length, with a sixth 
smaller spine close to one of them. At times 2 of the spines are reduced in size. 

Mueller (Vict. Nat., vii., 47) made the species a variety of B. Birchii F.v.M. 
and the type is labelled by him as B. Cornishiana, B. Birchii var. I havie ex- 
amined the type of B. CornisMcma (In the vicinity of the Field River, near the 
boundary line between Queensland and South Australia, W. H. Cornish, 1885) 
and, although it resembles B. Birchii F.v.M. in the woolly tomentum, obovate 
leaves and robust character, it differs considerably from that species in the nature 
of the fruiting perianth. The fruiting perianth of B. Cornishiana is much 
larger, with a broad flat top and with 6 horizontal spines more or less evenly 
placed. The fruiting perianth of B. Birchii is much smaller, more irregular, not 
nearly so broad at the top, and with only 5 vexy irregular spines. The spines 
of B. Birchii are also much longer than those of B. Cornishiana. 

Although, as I have previously pointed out, the type of B. Birchii does not 
absolutely agree with the specimens accepted by me as that species, yet it lies 
much closer to those species than to B. Cornishiana. In B. BircMi there are long 
unbranehed shoots with the flowers more crowded towards the apex. In B. Cor- 
nishiana these long shoots ane not so marked, and the flowers and fruits are more 
scattered and not so numerous. 

Bange.— In addition to the type we have two specimens collected at Ilpilla, 
and also what appears to be a form of this species with longer spines up to 10 
mm., and with not so broad a top to the fruiting perianth, from the upper Gas- 
coyne R., W.A. 

31. Bassia Birchii F.v.M. 

Anisacantha Birchii F.v.M., Fragm., viii., 163. — B. Birchii, First Census, 
1882; Icon. Austr. Salsol., PL Ixxii. (in part). 

This is probably the most unsatisfactorily defined species of the genus. It 
was described by MueUer from a plant collected on the Bowen Downs, Queens- 
land, by C. W. Birch. Dr. Laidlaw kindly sent over the type specimen and all 
material labelled B. Birchii in the Melbourne Herbarium. An examination of this 
material showed that there were 3 distinct species: 

(1). B. Birchii proper which was represented by a solitary specimen, the 
type. 

(2). B. Cornishiana which Mueller (Vic. Nat., vii., 47) made a variety of 
B. Birchii, but which is a quite distinct species. 

(3). A species which I have described as B. convexula, the difference from 
B. Birchii being detailed under the description of the former species. 

Although the type specimen was very imperfect, yet, in regard to the fruit- 
ing perianth, the general habit, tomentum. and shape of the leaves, it was ap- 
parent that it came very close to what I had regarded as a very robust foitn of 
B. quinquecuspis var. villosa. 

Specimens from the Queensland Herbarium similar to this robust form were 
all labelled B. Birchii. 

The Queensland and New South Wales specimens did not, however, altogether 
agree with the type of B. Birchii, differing chiefly in the length of the spines and 
mode of attachment of the fruiting perianth. The longest spines in the type are 
■under 4 mm. long, but in our specimens the spines are usually more than 4 mm. 
long and up to 15 mm. long. We have, howiever, specimens in which the spines 
are much reduced. 

In our specimens the fruiting perianth has a very oblique base and is very 
difficult to detach from the stem, whereas, in the type, the fruiting perianth has 



BY R, K. ANDERSOK. 345 

only a slightly oblique base and is more easily detaelied. Some of our specimens 
approach this condition. Comparison, however, of the fruiting perianth is 
liendered difficult by the absence of mature fruits in the type. Our specimens 
correspond with the type in the shape of the leaves, woolly tomentum, general 
habit, and the number and disposition of the spines. 

The type of B. Birchii is quite distinct from the type and other specimens 
of B. Cornishiana F.v.M. 

While recognising that there are slight differences between the type specimen 
of B. Birchii and the majority of our specimens I have decided to accept the 
latter as B. Birchii. 

Only two other courses are open: Firstly, to place B. Birchii with B. 
Cornishiana, making the latter a variety, and then to describe our specimens as 
a new species. But the differences between B. Birchii and B. Cornishiana are 
too decided to place them together (See under B. Cornishiana). Secondly, to 
uphold both B. Birchii and B. Cornishiana as distinct species and, in addition, to 
describe our specimens approaching B. Birchii as a new species. But the type 
specimen of B. Birchii only varies slightly from the Queensland and our own 
specimens, and in addition these specimens have been collected in and around 
the type locality. 

Should additional specimens show that the differences pointed out are con- 
stant, then, pierhaps, a new species could be described. 

Descriptive notes on B. Birchii F.v.M. — A robust branching under-shrub, the 
branches often sending off long secondary branches, or intricately branched, 
densely clothed in a thick woolly tomentum. Leaves obovate or broadly lanceo- 
late, up to 15 mm. long and 4-5 mm. wide, but usually smaller, densely hirsute. 
Tube of the fi-uiting perianth densely tomentose, very short, depressed or shortly 
turbinate, rarely more than 2 mm. long, with usually an oblique base. Spines 
unequal, 5, or 4 with one spine bifid, the longest spine up to 15 mm. long, but 
asually less. Often three of the spines are much shorter than the others, two 
being much reduced. Occasionally 3 of the 5 spines are long and more or less 
equal. The fruiting perianth is usually only detached with difficulty from the 
stem, tearing avv^ay with it a strip of the stem tissue. The seed is usually 
obliquely horizontal. 

The species is . commonly known as Woolerino Burr, Galvanised Burr, or 
Eoley-poley, and is a pest in certain areas, being difficult to eradicate owing to 
its spiny nature. 

The plate of B. Birchii (Icon. Austr. Salsol.) is a faithful reproduction of 
the type excepting the separate branch on the right-hand side which is B. Corn- 
ishiana I'.v.M. 

Affinity. — It resembles very closely a robust form of B. quinquecuspis var. 
villosa Benth., differing chiefly in the thicker, more woolly tomentum, the obo- 
vate leaves, and the stouter, more robust branches and leaves. The fruiting 
perianth and the position of the seed of the two are very similar, but the spines 
of B. Birchii are generally rather stouter, but no longer, than those of B. quinque- 
cuspis var. villosa. 

Although the species gives one the impression of being a very robust form 
of B. quinqiiecmpis var. villosa, it can readily be separated from that variety 
and its characters are constant. 

Mr. Mitchell of Walgett states that the Galvanised Burr {B. Birchii) seldom 
becomes detached from the ground as is the ease in the ordinary Roley-poley 
(B. quinquecuspis and the var. villosa). 



346 EEVISIOX OF AUSTRALIAN SPECIES OF BASSIA^ 

Mueller (Fragm.; viii., 163) draws attention to the distinction between this 
t^peeies and A. divaricata var. villosa. There is no var. villosa of A. divaricata, 
so probably this was a slip, Mueller having in mind B. quinquectispis var. villosa. 

liange. — This species is becoming fairly common in New South Wales and 
Queensland, and is liable to be a serious pest. I have seen about 10 specimens 
in the Queensland Herbarium representing Queensland localities. The following 
localities are represented in the National Herbarium: — 

Wee Waa, Gunnedah, Parkes, Narromine, Walgett, Wingham, and Jericho, 
Queensland. 

32. Bassia convexula, n.sp. (Plate xxxvi., A.-C.) 

Fruticulus sub-erectus, ramis tomentosis nonnunquam glabrescentibus cum 
ramulis longis erectis, foliis linearii-clavatis sessilibus acutis dense villosis 5-10 mm. 
longis (nonnunquam 15 mm.), floribus solitariis, perianthii fructiferi tube tomen- 
toso sub-gioboso vel turbinate sum apice convexo ad basin vix excavatam circ. 2 
mm. longo, limbo brevissimo, spinis 4 sub-divergentibus quarum una semper bifida 
(rarissime 5), spinis bifidis 1-li mm. longis, ceteris 2-4 mm. longis, infra medium 
tomentosis, semine horizontali. 

A more or less erect under-shrub, the brandies sending up long shoots up 
to 25 cm. long, the whole plant tomentose or villose, or the older branches in time 
becoming almost glabrous. Styles 2. Leaves linear-clavate, acute, and usually 
5-10 mm. long, but sometimes longer. 

The tube of the fruiting perianth is sub-globose or almost ovate-truncate, 
convex at its apex, tomentose, and slightly, or not at all, ribbed. Limb short. 
Fruiting pierianth with usually 4 somewhat ascending spines, one of which is 
always bifid, the bifid spines being much shorter than the others. Base of the 
tube slightly concave and somewhat circular in outline. 

A very distinct species which, in the past, has been confused with various 
species, principally B. quinquecuspis F.v.M., B. ecMnopsila F.v.M., and B. BircMi 
F.v.M. 

Although quite distinct from B. ecMnopsila^ it has been most commonly eon- 
fused with that species. 

It can be readily separated from B. echinopisila by the dense tomentose 
vestiture of the fruiting perianth, the number of spines, the shape of the tube 
and the horizontal seed, the two species having very little in common. 

Mueller, most probably, originated the mistake as he has figured both species 
in his plate of B. ecMnopsila (Icon. Aust. Salsol., Plate Ixix.). The extreme 
right-hand figure of series 6, and the top and lower left-hand figures of series 
7 are all B. convexula. The differences between the fruiting perianths can at 
once be seen from the figures. 

In addition there is in the National Herbarium a specimen from the Warrego 
River (E. Betche), labelled by Mr. Betche as having been determined by Mueller 
as B. ecMnopsila, the greater part of which is B. convexula^ with a small frag- 
ment of B. ecMnopsila. 

Mueller and others have also confused it with B. BircMi F.v.M. In the Re- 
port on the Elder Exploring Expedition (Trans. Roy. Soc. S.A., xvi., 1892, 346) 
Mueller refers to a "remarkably dwarf form, although copiously fruiting" of 
B. BircMi. This specimen is B. convexula. Although not altogether satisfied as 
to the true nature of B. BircMi, I am quite satisfied that it cannot be referred to 
this species. I have seen the type of B. BircMi and it is a much more robust 
plant, with a thicker, more woolly tomentum and broader leaves, which cannot 
be confounded with those of B. convexula. The shape of the fruiting perianth 
and the nature of the spines are also different. 



BY R. II. ANDERSON. 347 

B. convexula is readily separable from B. quinquecuspis F.v.M. by its 
easily detached, regularly-shaped fruits, the dense tomentum, the shape of the 
tube and the more erect, less intricately branched habit of the plant. 

Mange. — The species is widely distributed in the drier parts of Western 
Australia, South Australia, and New South Wales. The following localities are 
represented in the National Herbarium: — Girilambone, Poppy Mt. nr. Cobar, 
Coolabah, Broken Hill, Warrego R., Cobar, Paroo R. In addition I have seen 
some South Australian specimens sent over by Mr. J. M. Black. The only 
Western Australian locality I have seen is that recorded by the Elder Exploring 
Expedition, already quoted. 

33. Bassia parviflora, n.sp. (Plate xxxvi., H.-L.) 

Fruticulus sub-erectus, ramis gracilibus parce hirsutis, foliis lineari-clavatis 
sessilibus 4-8 mm. longis (rarissime 12 mm.) parce hirsutis, fioribus solitariis, 
perianthii fructiferi tubo plus minusve hirsutis subdepresso vel turbinato 1-H mm. 
longo cumt limbo brevissim.o, spinis 6, vel 5 quarum una semper bifida, 1-2^ mm. 
longis, semine horizontali. Approximat B. convexulae. 

A small, slender, twiggy plant v^ith intricate branches. Leaves linear clavate, 
small and scattered, rarely exceeding 8 mm. long. Flowers solitary in the axils 
of the leaves. Styles 2. Tube small, depressed or broad turbinate, flat on top, 
1-2 mm. long and bearing 6 small spines, or 5 with one spine bifid. Spines 
spreading horizontally or slightly inclined to the vertical. The leaves, branches, 
and fruits are more or less covered with a vestiture of long, somewhat thickened, 
translucent hairs. 

This species approaches closely to B. convexula but differs in 

(1) The fruiting perianth; it has not the domed apex so typical in B. con- 
vexula, and the spines spread more horizontally and are always one more in 
number than those of B. convexula. Very occasionally B. convexula has 6 spines, 
or 5 with one bifid, but then it is only in isolated instances in the one specimen. 

(2)_ The general habit, which is more twiggy than that of B. convexula, 
and in the smaller, more scattered leaves and fruits. The plant is weaker and 
not so robust. 

(3) Nature of the vestiture. In B. convexula the vestiture, particularly on 
the branches, consists of a fine cottony tomentum, that is to say, the individual 
hairs are weak, fine, whitish, and fairly short. The hairs of B. parviflora are 
of quite a different character, being longer, much stouter, straight, and trans- 
lucent. The difference in the vestiture is quite apparent when examined through 
a lens, and I have always found the differences to bei constant. 

Three of the four specimens I have placed under this species wene labelled 
B. echinopsila, but the species bears no relation to B. ^ecMnopsila, differing in the 
vestiture, shape of the fruiting perianth and the position of the seed. 

Range. — The four specimens I have placed under this species are from widely 
scattered localities in Victoria, South Australia and Western Australia. They 
are: — Mildura, Vic, Victoria Springs, Minnipa, S.A., and Doodlakine, W.A. 

34. Bassia costata, n.sp. (Plate xxxvi., D.-Gr.) 

Fruticulus sub-erectus, ramis parce tomentosis, fohis hirsutis lineari-subclavatis 
sessilibus 6-10 mm. longis, fioribus solitariis, perianthii fructiferi tubo subcylindrico 
parce hirsute demum glabrescente U-2 mm. longo, U mm. lato ad apicem planum 
et 1 mm. lato ad basin annulatam vix excavato. cum costa prominente ad spinam 
bifidam adscendente, limbo brevissimo, spinis 4, 2-4 mm. longis, quarum una semper 



348 REVISION OF AUSTRALIAN SPECIES OP BASSIA^ 

bifida, horizontalibus vel leviter recurvatis, semine horizontali. Approximat B. 
convexulae . 

60 miles N.E. of Camp 12, Northern Territory (G. F. Hill, 6/1911). Quoted 
as B. ecJiinopsila by Ewart and Davies (Flora of tbe Nortbern Territory, 1917, 
p. 95); Glen Ferdinand, Musgrave Ranges (S. A. White, 7/1914). 

A small erect plant, tbe branches sparingly tomentose, the leaves densely hir- 
sute. Leaves linear, slightly clavate, acute, 6-10 mm. long. 

The fruiting perianth is slightly hairy, but becoming almost or quite glabrous, 
flat on top, and is furnished with 4 spines, one of which is always deeply and 
regTilarly bifid forming two spines somewhat smaller than tbe other 3 spines. 
The two short bifid spines are united at their base to form a short, flat, broad 
projection of the fruiting perianth marked on the under surface by a prominent 
rib which continues down the tube. Tbe remainder of the tube is slightly ribbed 
or striate. The spines spread horizontally or are slightly recurved, the longer 
ones up to 4 mm. long. The tube narrows towards the circular, slightly hollowed 
base. Styles 2. Seed horizontal. 

This species is allied to B. convexula, but differs from that species in the 
glabrous, or almost glabrous fruiting perianth, in the flat top of the tube, in the 
more prominent ribbing of the tube beneath the bifid spines, and in the stouter 
horizontally spreading spines. The fruiting perianth of B. convexula is domed, 
tomentose, and the spines are more aeicular and ascending. 

35. Bassia microcarpa, n.sp. (Plate xxxvi., M.-Q.) 

Fruticulus ramosus, ramulis longis gracilibus pilosis, foliis sessilibus linearibus 
acutis 6-10 mm. longis dense pilosis, floribus solitariis, perianthii fructiferi tube 
depressus-cylindrico costato I mm. longo circ. 1 mm. lato glabro cum limbo 
brevissimo, spinis S quarum una semper bifida horizontalibus ad apicem valde 
recurvatis circ. 4 mm. longis, semine horizontali. 

Poison Creek, via Leonora, ¥/. Aust. (A. McGregor, 9/1909.) ; the type. 

A very distinct species, easily separated from other members of the genus 
by the long, soft, dense hairs on the leaves and the numerous very small fruits. 
Styles 2, connate at the base. The tube of the fruiting perianth is very shortly 
cybndrical, under 1 mm. long and 1-lJ mm. in width, glabrous, and ribbed verti- 
cally; slightly hollow at the base and marked by 5 distinct perforations. Spines 
5, one of which is bifid, equal and placed at regular intervals around the fruiting 
perianth, about J mm. long, spreading horizontally, but sharply hooked at the 
ends, the tips often being much darker in colour. At times the spines appear to 
be united by a very narrow rim or expansion of the fruiting perianth, showing 
an affinity with the Maireana section. 

I have seen only the one specimen. At first I believed that this jolant might 
be B. micranthera (Benth.) F.v.M. The type of this species was Drummond's 
n. 253, Fourth Coll. (W.A.), and Mueller (Fragm. vii., 12) refers it with doubt 
to Kochia villosa 

Bentham (B.FL, v., 181) describes it as Enchylaena 1 micranthera, making 
reference to Muellei-'s opinion, but stating that he can see no trace of any wing 
or transverse prominence to the perianth. He also describes the perianth as 
"already much enlarged after flowering with the fruit nearly ripe, globular, 
glabrous or hairy especially the lobes, smooth and rather thick scarcely f line 
diameter; lobes 5 short, obtuse, connivent over the fruit." He can find no trace 
of appendages, and accordingly will not allow it to remain under KocMa, but 
transfers it provisionally to Enchylaena. 



BY E. H. ANDERSON". 349 

This description does not agree in any sense with the perianth of B. niicro- 
carpa which is furnished with 5 or 6 distinct spines which are quite apparent 
even in the flowering stage. Altliough Mueller later transfeiTed Enchylaena 
rmcr anther a to the genus Bassia, it does not follow that he observed spines on 
the fruiting perianth, as he included several non-spined species in that genus, 
e.g., B. enckylaenoides and B. carnosa. 

I obtained the type specimen of B. miicr anther a from the Melbourne Her- 
barium, but this had no flowers or fruits. A specimen from Althorpe, S.A. (Dr. 
R. S. Rogers, 9/1907) matched the type very closely, but this also had neither 
flowers or fruits. The vestiture of the type and that of B. micrGcarpa are prac- 
tically identical, but this is not a strong point, as the vestiture is also identical 
with that of B. hrachyptera (F.v.M.). 

As Bentham's description of the perianth of E. micranthera is so clear and 
leaves no room for doubt, I could not refer the specimen of B. mcrocarpa to it, 
although both agree in the nature of the vestiture and the smallness of the fruits. 
Two minor differences between B. microcarpa and B. micranthera lie in the num- 
ber of the styles and the general habit of the plant. Bieintham gives the number 
of styles of E. micranthera as 3, whereas in B. mcrocarpa I have only found 2. 
The secondary branches of B. microcarpa are apparently much longer and more 
slender than those of B. micranthera. Owing to the nature of the perianth and 
the absence of spines in B. micranthera (Benth.) F.v.M., this species cannot be 
retained in the genus Bassia, and perhaps is best placed under Enchylaena where 
Bentham originally placed it. The flowers and ripe fruits of the species, how- 
ever, are necessary to determine its position satisfactorily. 

36. Bassia bjflora (R.Br.) F.v.M. 

Sclerolaena biflora R.Br., Prodr., 410. — Dissocarpus hiflorus F.v.M., Trans. 
Phil. Inst. Vict., ii., 75. — B. biflora, Fragm., xii., 1882, 12; Icon. Austr. Salsol., 
PL Ixi. 

A procumbent shrub clothed in a white woolly vestiture. Leaves linear, 
acute, up to 20 mm. long, but usually 8-15 mm. The connate fruiting perianths 
are usually without spines, but occasionally short spines or tubercles are present. 
In a specimen from Mt. Sturt Station, Broken Hill District (A. Morris) there 
are 1-3 well developed spines, 1-2 mm. long. These are usually 2 in number, one 
being on either side of the connate perianths. 

The spiny form varies slightly in the nature of the fruiting perianths. 
These, in the typical non-spined form, are horizontally divergent from the common 
base, but in the spined form one is almost vertical or inclined upwards, the other 
diverging horizontally. 

Bange. — In addition to the specimen already quoted the following localities 
are represented in the National Herbarium: — Murray R., Lachlan R., Barrier 
Range, Mt. Lyndhurst, S.A. 

Var. CEPHALOCARPUM F.v.M. (Fiagm., viii., 38). 

Fruiting perianths 3-7 in the head. The vestiture of the variety is usually 
not so dense as that of the typical species, but no sharp line of demarcation is 
possible between the variety and the species. For example 2-flowered and 3- 
flowered heads may be present on the one plant. 

The variety is readily distinguished from B. paradoxa (R.Br.) F.v.M. by the 
almost complete absence of spines. 



350 REVISION OF AUSTRALIAN SPECIES OF BASSIA^ 

The following localities represent the variety in the National Herbarium: — 
Bourke Dist., Brewarrina, and Mooculta Siding. 

37. Bassia paradoxa (R.Br.) F.v.M. 

Sclerolaena paradoxa R.Br., Prodr., 1810, 410. — Chenolea paradoxa F.v.M,, 
Fragm., x., 91. — B. paradoxa, First Census, 1882; J. Hook., Icon. PL, 1076; 
F.v.M., Icon. Austral. Salsol., PI. Ixii. 

Notes additional to the description. — A decumbent shrub densely clothed in 
a loose tomentum. The fruiting perianths are connate into a hard mass of 8-16 
individuals, the whole measuring up to 15 mm. or more in diameter. J. M. Black 
(Trans. Roj^ Soe. S.A., 1916, 459) found the number of connate fruits in the 
head to be 8, each being 2 — 5-spined. On dissecting the heads, I have found as 
many as 16 fruits. 

As far as I have observed in the specimens in the National Herbarium, the 
heads are alv/ays spiny, the spines being up to 5 mm. long, but usually shorter. 
The leaves are thick, and up to 15-20 mm. long by about 3 mm. in width, but 
usually smaller. The seed is placed horizontally. 

Var. LATiFOLiA J. M. Black (Trans. Roy. Soe. S.A., xlvi., 1922, 567) . 

This variety is distinguished by the much broader leaves (5-8 mm. in width), 
and by the spines being reduced to 5 short obtuse and rather soft horns. From 
the limited number of available specimens, Mr. Black is unable to say whether 
this characteristic of the fruit appendages is constant. 

Mueller (Fragm., vii., 11) pointed out that the leaves of B. paradoxa were 
sometimes 3 lines (about 6 mm.) in width, so probably he had in mind this variety. 

Range. — This species enjoys a fairly wide range in the drier parts of Queens- 
land, New South "Wales, Victoria, South Australia, and Western Australia. 

38. Bassia spinosa Ewart and Davies. 

Fl. N. Terr., 1917, p. 95, PI. xii. 

I have seen no specimens of this species but, judging by the description and 
plate, it is a distinct species, approaching B. paradoxa (R.Br.) F.v.M., but 
differing from that species in the fewer flowers in the head and iri the stronger, 
divided spines. 

39. Bassia tridens F.v.M. 

Fragm. xii., 1882, 12; Icon. Austr. Salsol., 1889, PI. Ixiv. 

A branching under-shrub, at first very slightly hirsute, but soon becoming 
glabrous, except the axils of the leaves, base of the fruiting perianth, and the 
flowers. Leaves linear, semi-terete, 5-12 mm. long; tube of the fruiting perianth 
very shortly cylindrical, somewhat dilated towards the much or slightly hollow 
base, 1-2 ram. long. Fruiting perianth flat on top, with 3 horizontally spreading, 
broad, flattened appendages, one of which is bifid. These appendages resemble 
much flattened spines and, at their base, are 2-2^ mm. across, but taper to a 
sharp point. Each flattened appendage is about 5-6 mm. long, much hardened, 
and on the under surface is marked by a longitudinal rib which continues down 
the perianth tube. The seed is hoinzontal. 

This species is quite distinct from other members of the genus by reason of 
the broad flattened appendages, which may be taken to represent flattened spines. 
It certainly deserves a sectional, if not a generic, distinction. 



BY B. H. ANDERSON. 351 

Seetionally it comes nearest the Maireana Section, but differs in the regailar, 
well defined spines and the absence of the horizontal wing expansion. 

The only specimen I have seen is the type, Gaseoyne R. (J. Forrest). 

40. Bassia Luehmanni F.v.M. 

Icon. Austr. SalsoL, 1889, PI. Ixx.; Viet. Nat., vii., 1890, 47 and Bot. 
Centralb., xliii., 1890, 371. 

A rather dwarf under-shrub, more or less densely hirsute, or the leaves 
sparingly so. Leaves rhomboid in the upper part and narrowing into a fairly 
long petiole-like base. The tube of the fruiting perianth is almost flattened, very 
short, being barely -} mm. long, and is furnished with a wide hardened expansion 
broken up very irregularly into spines or spine-like segments which often are, in 
turn, again broken up or bifid. The fruiting perianths are densely clustered 
along the branches and are sparingly hirsute. Seed horizontal. 

This species is readily separated from all other Bassias by the very short 
tube, the broad expansion broken up into spines, and the peculiar leaves which 
are found in this species alone. The broken expansion shows an affinity with 
B. stelligera F.v.M., but the insignificant tube is very different. 

The only specimen I have seen is the type, Finfee R. (Rev. W. T. Schwarz, 
1886) . 

41. Bassia stelligera F.v.M. 

Maireana stelligera F.v.M., Fragm., i., 139. — EcMnopisilon stelligeriis, Fragm., 
vii., 13. — Kochia stelligera F.v.M., Fragm., vii., 13 (inferred) and B.FL, v., 
189. — B. stelligera, F.v.M., MS. Third Census; Icon. Austr. SalsoL, PI. Ixviii. 

Notes additional to the description. — A procumbent under-shrub clothed in a 
dense, woolly, white tomentum, but the stems at times tending to become glabrous. 
Tube of the fruiting perianth globose, usually smooth and with a thin horizontal 
expansion which is broken up into spines. These spines are sometimes 15 in 
number, but usually about 8-12, often bifid or with secondary spines, and about 
1 mm. long. The unbroken section of the expansion is ^-1 mm. or even more in 
width. At times the expansion is only partially broken up, the spines being re- 
presented by irregular denticulations, marked by ribs on the under side. The 
leaves are up to 20 mm. long, but usually are much shorter. 

This species, together with the other members of the section, shows the 
transition from the genus Bassia to the genus Kochia. 

The following localities are represented in the National Herbarium: — • 
Victoria: Gannawarra; N.S.W^: Berrawinia Downs, Nevertire, Namoi R., Wan- 
ganella, Cobar, Warrakoo, Coonamble. 

42. Bassia brachyptera (F.v.M.), n. comb. 

Kochia hrachyptera, Sec. Gen. Rep., 1854, 15; Icon. Austr. SalsoL, PL lix. — 
Sclerochlamys hrachyptera F.v.M., Trans. Phil. Inst. Viet., ii., 76. — Echinopsilon 
brachyptera, Fragm., vii., 13. 

Mueller (Fragm., xii., 15) pointed out that this species and Kochia stelligera 
F.v.M. merged into the genus Bassia. In his M.S. of the Third Census he placed 
Kochia stelligera under Bassia and also figured it under that genus in Icon. Austr. 
SalsoL, but he preferred to allow Kochia hrachyptera to remain under Kochia. 
Its affinities, however, lie with Bassia rather than with Kochia, although, like 



352 " REVISION OF AUSTRALIAN SPECIES OP BASSIA^ 

other members of the Maireana Section, it is really a transition form between the 
two genera. The fruiting perianth is much hardened, like a Bassia, and the ex- 
pansion is hardened and marked by 5 very short spines. 

Mueller (Vict. Nat., vii., 1891, 66) suggested that the three species, K. 
hrachyptera, K. stelligera, and K. ciliata be excluded from both Kochia and 
Bassia., forming a separate generic group. For this he proposed the name 
Scleroclilamys, but added that the three species might be retained under Bassia if 
a wider definition were given to that genus. Kochia ciliata, however, undoubtedly 
belongs to the genus Kochia as the membraneous expansion is quite decided. 

Notes on B. brachyptera (F.v.M.), n.comh. — Leaves and branches densely 
invested with long hairs, the older branches at times almost glabrous. Leaves 
usually 10-15 mm. long. The fruiting perianth is glabrous, the tube short, about 
li mm. long and tapering to a circular base with 5 small perforations. The tube 
is usually marked by 10 longitudinal ribs, 5 of which are produced into short re- 
curved spines, united by the hardened expansion. The fruiting perianth is 2^-4 
mm. in diameter at the top, including the expansion, the fliat top being marked 
by 5 radiating ribs running out into the spines. The very short spines appear 
to be a prolongation of the ribbing of the tube, or as acute somewhat produced 
corners to the flat top of the fruiting perianth. 

The presence of the spines although much reduced and not typical of the 
genus, and the hard nature of the expansion indicate its closer afi&nity with the 
genus Bassia. Specifically its nearest ally is B. microcarpa, n.sp., a species which 
undoubtedly is a Bassia. I have placed it in the section Maireana because of the 
expansion of the fruiting perianth, but the nature of the spines and ribbing , is 
quite different from the other members of that section. Altogether, it is a difficult 
species to place satisfactorily. 

Bange. — N.S. Wales : Wilcannia, Yandama, Brewarrina, Condobolin, Warrego 
K., Bourke, Darling R., Mt. Browne ; Vic. : Lake Hindmai-sh, Murray R. ; S. Aust. : 
Patawurta, Flinders Range, Moolooloo Station between Beltana and Blinman, Mt. 
L^mdhurst, Ediowie. 

Species excluded from the genus Bassia, or imperfectly known. 
Bassia longifolia W. V. Fitzgerald, Muell. Bot. Soc, 1904, 32. 

We have the tj'pe specimens of this species (Nannine, W.A., Sept., 1903) 
and they are identical Avith Kochia fimbriolata F.v.M. The latter was described 
in Fragm., ix., 75, and a plate is given in Icon. Austr. Salsol., both description 
and figure agreeing in every parfieular with the type specimen of Bassia longi- 
folia W. V. Fitz. The short hardened tube with the 5, scarcely membraneous or 
somewhat hardened, horizontal wings, the 5 erect appendages, and the long hir- 
sute leaves are the chief features of the species. 

Fitzgerald, in his description of B. longifolia, remarks that it approaches 
K. fimbriolata, but does not point out wherein it differs, and the possibility is 
that some confusion existed as to the true nature of the latter species. The 
possession of a hardened tube, although not very common in the genus Kochia, 
does not exclude a species from that genus, and the 5 wings are a characteristic 
generic feature. 

Bassia longifolia W. V. Fitz. must, therefore, be reduced to a synonym of 
Kochia fimbriolata F.v.M. 



BY B. H. ANDERSON^. 353 

Bassia carnosa (Moq.) T.v.M., First Census, 1882. 

EcMnopsilon 1 carnosus Moq., D.C. Prod., xiii. (ii.), 136. — Trichinium car- 
nosum Moq. in herb. Hook. — Chenolea carnosa Benth., B.Fl., v., 1870, 190. — 
Bassia lanuginosa C. T. White, Q'ld. Agrie. Joum., xv., 1921, 216. 

0. Paulsen (Dansk. Botanisk Arkiv., ii. (8), 1918, 58) points out that, as 
the perianth of this species is wingless and spineless, the species must be included 
under Chenolea. I have examined the fruiting perianths of this species which 
are concealed under dense white wool, and have always found them to be fur- 
nished with a definite horizontal wing or expansion. This wing is continuous, 
semi-membraneous and transparent, or thickened and opaque, and is usually 
about 1 mm. or occasionally more in width. The fruiting perianth is hardened. 

The species, therefore, must be placed in the genus Kochia, as it has the 
two main characteristics of that genus, namely, wing expansion and altered 
perianth. The fruiting perianth of Chenolea, on the other hand, is unaltered, 
and free from any appendages or wing expansion. I, therefore, propose the 
name Kochia carnosa (Moq.), n. comb., for the species. 

Synonym. — Bassia lanuginosa C. T. White. 

Through the Director, Mr. Maiden, I have corresponded with Mr. White and 
he agrees that this species is synonymous with Kochia carnosa (Moq.), n. comb. 

The nature of the fruiting perianth, the vestiture of the leaves and fruits, 
and the peculiar arrangement of the floral leaves are identical in the two sipecies. 

The distribution of the two species is certainly peculiar. KocMa carnosa is 
a Western Australian species, whereas Mr. White's specimen comes from the 
Mayne R., Queensland. As it is apparently an isolated patch, it may possibly 
have been introduced, but in all events geographical separation is not sufifleient 
ground for specific distinction. 

The only differences between the Queensland specimens and the Western 
Australian species appear to be the slightly greater size of the leaves of the former, 
a slightly greater development of the wool about the fruiting perianth, and the 
flowers and fruits being in a looser arrangement. On these points the Queensland 
specimens, especially if added to by similar specimens from other localities, may 
possibly be distinguished as a variety. 

Bassia hostilis Diels, Engl. Bot. Jahrb., xxxv., 1904, 185. 

I have seen no specimens of this species and no figure is given with the 
description, and I therefore hesitate to express a definite opinion of the species. 
But on the description alone, I am unable to separate it from Bassia divaricata 
(R.Br.) F.V.M., the features of both species apparently agreeing in every par- 
ticular. Diels quotes B. hostilis as "approaching B. divaricata in some respects, 
e.g., dioecious flowers and spines united at the base," but omits to enumerate the 
points of difference. Being therefore imperfectly known to me, I was unable 
to include the species in the revision. 

Bassia suadacea F.v.M., Viet. Nat., vii., 1890, 42 (nomen nudum.) 

The name appears in the minutes as that of a species included in a paper 
read by Luehmann on behalf of Mueller, but neither the name nor description is 
included in the paper, so apparently it was suppressed by Mueller. I can find 
no further trace of it. 



354 REVISION OF AUSTRALIAN SPECIES OF BASSIA_, 

Kentropsis cornuta F.v.M., Sec. Gren. Report, 1854, 15. 

Other species included in the genus Kentropsis were later transferred to 
BaSsia. The above species, however, appears as a nomen nudum in the See. 
Gen. Report, and I can find no subsequent reference to it. 

Sclerolaena coriacea Moq. in D.C., 13 (ii.), 123. 

Bentham (B.FL, v., 1870, 194) remarks that this species is unknown to him 
and, the perianth being devoid of spines, may not be a true congener. I have 
found no further reference to the species. 

Bassia enchylaenoides F.v.M. and Bassia tricornis (Benth.) F.v.M. 

Owing to the character of the fruiting perianth of these two species, they 
must be excluded from the genus, and apparently do not come within the limits 
of the allied genera. I intend dealing with them separately in a subsequent 
paper. 

Bassia micranthera (Benth.) F.v.M. 

I have dealt with this species under B. microcarpa, n.sp. 



EXPLANATION OF PLATES XXXIV.-XXXVI. 

Plate xxxiv. 
Bassia patenticuspis, n.sp. 
A. Branch. B. Fruiting perianth. C. Vertical section of fruiting perianth 
showing position of seed. 

Barrier Ranges. Miss M. I. Collins, No. 16, Aug., 1921. 

Bassia ohliqwicuspis, n.sp. 

D. Branch. E. Top view of fruiting perianth. F. Basal and lateral view of 
fruiting perianth. G. Vertical section of fruiting perianth showing position of 
seed. 

Port Augusta, S.A. J. M. Black, Sept., 1920. 

Bassia parallelicu^pis, n.sp. 
_H. Branch. I. Side view of fruiting perianth. K. Basal view of fruiting 
perianth. L. Vertical section of fruiting perianth showing position of seed. 
Mt. Lyndhurst, S.A. Max Koch, Sept., 1899. 

Plate XXXV. 
Bassia tubata, n.sp. 
A. Portion of plant. B. Fruiting perianth. C. Vertical section of fruiting 
perianth showing position of seed. 
Coonamble. E. Breakwell, 1918. 

Bassia intricata, n.sp. 
p. Branch. E. Fruiting perianths in situ on the stem. F. Vertical section 
of the fruiting perianth showing position of the seed. 
Mutooroo, S.A. A. Morris, No. 635, Aug., 1921. 

Plate xxxvi. 
Bassia convexula, n.sp. 
A. Branch. 13. Fruiting perianth. C. Section of the fruiting perianth show- 
ing position of the seed. 

Cobar. L. Abrahams, Aug., 1911. 



BY R. H. ANDERSON. 355 

Bassia costata, n.sp. 
D. Portion of plant. E, F. Side and top view of fruiting perianth. G. Ver- 
tical section of the fruiting perianth showing position of the seed. 
Sixty miles N.E. C.2. N. Territory. G. F. Hill, June, 1911. 

Bassia parviflora, n.sp. 
H. Branch. I, K. Side and top view of fruiting perianth. L. Vertical sec- 
tion of fruiting perianth showing position of the seed. 
Minnipa, IS.A. J. M. Black, Nov., 1915. 

Bassia microcarpa, n.sp. ^ 

M. Portion of plant. N. Side viev; of fruiting perianth. O. Top view of 

fruiting perianth. P. Bottom view of fruiting perianth. Q. Section of fruiting 

perianth showing position of the seed. 

Poison Creek, via Leonora, W.A. A. A. McGregor, Sept., 1909. 



356 



STUDIES IN PLANT PIGMENTS. 

Part i. — The Yellow Colouring Matter of the Acacias. 

By James M. Petrie_, D.Sc, F.I.C, Linnean Macleay Fellow of the Society in 

Biochemistry. 

(From the Physiological Laboratory of the University of Sydney.) 

(Five Text-figures.) 

[Read 29th August, 1923.] 



Introduction. 



Our knowledge of plant colours has been traced back to the old civilisations 
of Persia and India. It is closely associated with the art of dyeing as practised 
by these ancient nations. 

Though from nearly every plant may be extracted some potential colour sub- 
stance which, if combined with certain chemical salts, will produce a dye, there 
are very few plants found to contain an actual dye-stuff. The latter may be 
exemplified by indigo blue, the brown stain of walnut fruits, and the yellow and 
brown pigments contained in some lichens. 

The pigments, so widely distributed in the vegetable kingdom, are found 
however in greater concentration in certain specific plants. Some of these con- 
tain yellow colour substances which have been used for centuries, such as alkanet 
of the Romans, turmeric of India and China, weld used by the Gauls before their 
conquest by Caesar, safflower of the Egj^ptians, fustic, and quercitron. 

Our chemical knowledge of these yellow j)lant pigments begins with the 
study of quereetin by Herzig in 1884, and recent research has disclosed the fact 
that many of them belong to the flavone group of organic compounds; indeed, 
as Perkin has said, of all the natural dyes these yellow flavone piginents are the 
most widely distributed in nature. 

Plant pigments are considered biochemically in two distinct grouiDS — the plas- 
tid pigments and the water-soluble pigments. In the former the molecule of the 
colour substance is combined in some obscure chemical way with the other com- 
ponents of the complex colloidal molecule of protoplasm; in this group chloro- 
phyll, carotin, and xanthophyll are well known. Of the water-soluble pigments 
found in solution in the cell-sap, the yellow coloured substances are mostly flavone 
and xanthone pigments, while the blue, red and purple colours of flowers are 
called anthocyanins. 




by j. m. peteie. 

Experimental. 
I. Acacia discolor. 
II. Acacia lini folia. 

III. Acacia decurrens, var. mollis. 

IV. Acac^a longifolia. 

The Aeaeias aft'ord us the best available material for the study of the yellow 
pigments of plants. They constitute, with the Eucalypts, a large part of the 
unique flora of Australia; more than 300 endemic species are distributed through- 
out the continent, and in all seasons of the year one or more is to be seen in 
flower. The beautifvil florescent masses consist largely of dense crowds of 
stamens, petaloid and unprotected, and their colour ranges from palest cream to 
deepest golden yellow. 

Hitherto nothing was known of the chemical nature of these golden wattle- 
blossoms, or of the substances to which they owed their beauty. In this investi- 
gation the yellow pigments have been extracted, and their properties studied. 

The Materia^. — Four typical examples of Acacias were chosen for this in- 
v^tigation: two of these possess the characteristic pinnate mimosa leaves, and 
two have their leaves replaced by phyllodes in response to their xerophytic en- 
vironment. The flowers were collected in quantity at the period of maximum 
inflorescence, in the neighbourhood of Sydney and the Blue Mountains. 

General Method of Extraction. — As soon as possible after collecting, the fresh 
flowers were immersed in small amounts at a time in boiling water, in order to 
destroy the enzymes. The boiling was continued for some hours, after which 
the material was pressed, and again extracted until most of the soluble colouring 
matter was removed. The large volume of fluid was evaporated to about 2 litres, 
after which the proteins were coagulated with acetic acid and removed. 

The aqueous fluid was then precipitated by normal and basic lead acetate; 
the precipitates contained the colouring matter and tannins. The lead was re- 
movecl, and the pigments were hydrolysed by boiling with dilute sulphuric acid, 
during which operation the glucosides were decomposed into yellow pigment and 
sugars. The pigment was next extracted by agitation with ether, and the re- 
maining aqueous fluid was reserved for the sugar. On distilling the ethereal 
extract the colouring matter was obtained in the crude state, and was then 
purifled by a series of fractional crystallisations. 

I. Acacia discolor. 

The flowers of A. discolor, the sunshine wattle, were collected at Sutherland 
and Hornsby in the month of April. 

Extraction of the Soluble Yelloio Pigraent. — Hot aqueous extracts were made, 
which after concentration were agitated with ether. The absence of yellow pig- 
ment from this ethereal extract showed that no free flavone existed in the plant, 
and that the pigment was present wholly in the form of a glucoside. The latter 
was precipitated by lead acetate, and after removal of the lead from the precipi- 
tate, it was submitted to acid hydrolysis. This operation resulted in the for- 
mation of (1) a black resinous substance which was insoluble in the hot fluid; 
(2) a deep brick-red deposit which formed on cooling, and was removed by de- 
cantation; (3) a bright red-coloured solution. 

The second and third of these were completely extracted with ether, which 
gave a deep yellow solution of the pigment.- The last few extractions were brown 
coloured and yielded on evaporation no yellow pigment, but a pale brown 
amorphous substance which possessed an astringent taste and gave the reactions 
of a tannin. 



358 STUDIES IN PLANT PIGMENTS, i., 

The filtrate from the lead acetate precipitate obtained above was treated with 
basic lead acetate, when a further precipitation occurred. The lemon-yellow 
coloured deposit was hydrolysed in the same manner as the former and yielded 
a bright yellow ethereal solution. 

These ethereal extracts were concentrated and then agitated with successive 
small amounts of ammonium and sodium carbonates. The yellow colouring matter 
passed into both alkaline solutions from which it was reprecipitated by pouring 
into acid. 

The ammonium carbonate extract at first was green in colour; in half a 
minute it became red, and soon changed to purple and violet. These coloured 
solutions gave the reactions of anthocyanins. The yellow precipitates obtained 
from the acid solutions consisted of crude colouring matter. 

Purification of the crude yellow pigment. — The yellow crystalline deposits 
contained a red colouring matter and an amorphous brown substance intermixed. 
By dissolving in hot alcohol a deep red solution resulted, and on carefully diluting 
this vrith hot water a point was reached when the yellow pigment could be 
crystallised out alone. When recrystallised in this way a number of times, the 
filtrate at last ceased to be red, and a mass of lemon-yellow, needle-shaped 
crystals was obtained. These possessed an almost constant melting point, show- 
ing that only one substance was present, and were submitted to a rigid investi- 
gation to determine their properties and constitution. 

Products obtained from the above operations. — 1,200 grams of fresh flowers, 
or 500 gTams dried (at 100°C.) material gave: — 

24 g-ms. black resinous substance. 

32 gms. of brick-red deposit after hydrolysis, consisting of phlobaphenes. 

2.3 gms. of yellow pigment from the lead acetate pptn. 

0.1 gms. of yellow pigment from the basic lead acetate pptn. 

(2.4 g-ms. of yellow pigment or 0.2% of the fresh flowers). 

A solution containing the glucosidal sugars. 

A brownish yellow amorphous substance from the mother-liquors after 
crystallisation of the yellow pigment. 

Deep crimson coloured aqueous solutions exhibiting the properties of antho- 
cyanins. 

A viscid oil of strong acrid odour, obtained from the ether extract after the 
yellow pigment had been removed by alkali. 

(a.) The soluble yellow pigmient. 

In order to complete the purification of the yellow substance the whole was 
converted into its acetyl derivative!, by boiling with acetic anhydride and sodium 
acetate. The compound separated from water as a brown crystalline mass, and 
after purifying with charcoal and recrystallising many times from hot alcohol, a 
pure white product was obtained. The acetyl derivative possessed a constant 
melting point, and was afterwards quantitatively hydrolysed. The yellow pigment 
obtained in this way was now assumed to be pure. 

(1) Properties. — The substance appeared under the microscope as beautiful 
masses of crystals, fine hair-like needles of a pale lemon-yellow colour, and 
possessing a melting