APPENDIX IV 



331 



pale glandular dots, usually with the appearance of two grooves and a semi- 

 circular ridge between them. The opposite surface nearly round. When 

 quite fresh, the phyllode is doubtless sub-terete, with the two blunt margins 

 slightly incurved. No gland observed at base. 



Flowers in heads, 5-merous. The flower heads are clustered in the axils of 

 the phyllodes, 4 or 5 springing from stipules surrounding the pedicels at the base. 



Bracts with long pointed heads and thin stipes, standing out like prickles 

 beyond the petals in the unopened flower-head. 



Calyx very narrow spathulate, with a few hairs, about half the length of 

 the corolla. 



Petals united about half way up but separating at a touch, glabrous, with 

 nerve. 



Pistil smooth. Pods and seeds not seen. 



This interesting species belongs to the series CaZam«/orme5,sectionUniner\'es. 

 I dedicate it in honour of Mr. Charles Ernest Froderick Allen, Curator of the 

 Botanic Garden, Darwin, now at the front, who has done excellent work in 

 collecting species of this genus. 



Its closest affinity appears to be with A. praelongata, F. v. M., from which 

 it differs in the greater length of the phyllodes of that species, and also in their 

 comparative flatness with midrib and visible lateral nerves. 



The protruding bracts of the new species give it a different appearance to 

 that of A. praelongata, but the structure of the flowers is not very different. 



The flowers of the new species are in chisters, and not racemose as in A. 

 praelongata. 



When pods are available a further statement can be made in regard to the 

 position of the species. 



The prickly bracts are also observable in A. Simsii, A. Cunn., and ^4. 

 homoclada, F. v. M. 



I think it very possible that A. Alleniana has been passed over as A. 

 juncifolia* the unsatisfactoriness of whose record for the Northern Territory 

 I have already drawn attention to. I only know A . juncifolia from Queensland 

 and New South Wales at present, and proceed to compare A. juncifolia and A. 

 Alleniana, using for the purpose a specimen of the former species from the 

 Balonne River, Queensland, which tradition in the National Herbarium, 

 Sydney, says was collected by Mitchell, and it is probably a co-type. 



A. juncifolia, Benth. A. Alleniana, n. sp. 



Phyllodes shorter (up to 17 cm.), terete Phyllodes longer (up to 23 cm.), terete, 



or semi-terete, bent near base, where there is no nerve visible, gland absent, long wrinkled 



a gland, nerve visible, short wrinkled attach- attachment, 

 ment to the stem. 



Flower heads usually in pairs but some- Flower heads as many as 5 together 



times in threes. Peduncles short. in the axils of the phyllodes. Peduncles 



very long and slender. 



The bract on the pedxmcle figured at 2 No bract seen, 



of Mueller's Iconography Plate of A. 

 juncifolia is not present in this specimen. 



Calyx irregularly divided to the base, Calyx divided to the base, narrow, 



glabrous, but rugose or pappilose. It has ciliate at the top. 

 no hairs. 



Floral bracts shorter than the buds, Floral bracts much exceeding the buds 



hardly visible in the flower-head. in length, giving a pricldy appearance to the 



flower-head. 



* This is no longer a matter for surmise. See under A. juncifolia, p. 320, where I show 

 that a specimen from Kew ilabelted juncifolia), received at the Queensland Herbariiun, 

 turns out to be A. Adleniana. 



