PLANT MORPHOGENESIS FOR SCIENTIFIC MANAGEMENT OF RANGE RESOURCES 



151 



to 1 m. in height and attains densities in excess 

 of 12,000 bushes per hectare on an estimated 2-3 

 million hectares of rangeland. The leaves are un- 

 palatable to domestic stock except in the more 

 xeric habitats where it may be eaten during pro- 

 tracted droughts. By contrast, the plant flowers 

 profusely after rain and if stocking rate is high 

 the flowers are readily eaten by sheep. This ob- 

 servation suggests that stock numbers may be 

 manipulated to control the species by preventing 

 regeneration from seed. 



A detailed study of the ecology of Eremophila 

 gilesii is currently being undertaken by staff of 

 the Charleville Pastoral Laboratory and prelim- 

 inary results are presented here. 



Of the weed control measures listed by Kling- 

 man (17), only grazing management, biological 

 control, and fire appear to have application in 

 semiarid Queensland. Lands with woody weed 

 problems in this area are currently valued at 

 less than $A4 per hectare ($A= Australian dol- 

 lar = approximately $1.25 U.S.). Development of 

 management techniques and biological control 

 measures necessarily call for long term research 

 programs. Use of fire as a controlling agent is 

 dependent on the production of enough herbage 

 within the stand to fuel a burn of sufficient in- 

 tensity to kill the weed. In southwest Queens- 

 land, standing crops of grasses in areas where 

 woody plant infestation is high seldom reach 

 100g./m. 2 and are commonly much lower (8). 

 Moreover the buildup of fuel over several years, 

 as advocated by West (4£) on South African veld, 

 is prevented in Queensland by the depredations 

 of grass harvesting termites (38). Thus, ecologi- 

 cal control measures (including grazing manage- 

 ment) seem at present the only practical means 

 of reducing shrub densities in southwest Queens- 

 land. There is general agreement among Aus- 

 tralian workers on the need to look at the ecologi- 

 cal system as a whole rather than at individual 

 species (10, 21, 22). However, studies in which 

 this principle is adopted are rare. Moore (23) 

 pointed out that, in semiarid shrub woodlands, 

 removing one woody species may increase the 

 density of others. 



Distribution 



A distribution map for Eremophila gilesii was 

 constructed from information supplied by her- 



baria of mainland Australia (excluding Victo- 

 rio). This map (fig. 1) indicates the climatic 

 adaptability and range of the plant. The distri- 

 bution closely follows that of Acacia aneura (12). 

 This is not surprising in view of the highly sig- 

 nificant associations found between A. aneura 

 and Eremophila spp. by Burrows and Beale (3) . 

 Nevertheless the plant is a problem only in south- 

 west Queensland, the most mesic part of its geo- 

 graphical range. 



An examination of the climatic data presented 

 by Perry (31) showed that the species is largely 

 confined to areas receiving between 60 and 70 

 percent of the annual rainfall during the summer 

 months (October-March). All areas where it oc- 

 curs receive significant cool season rains. Temper- 

 atures experienced throughout the range are very 

 similar. For Charleville, mean monthly maximum 

 and minimum temperatures for the hottest month 

 are 36° and 22° C, respectively. Corresponding 

 temperatures for the coldest month are 20° and 

 4°. 



Experiments 



Methods 



Temporal changes in species density were re- 

 corded in small enclosures (40 m. by 10 m.) 

 placed on two spatially distant stands of Eremo- 

 phila gilesii, during 1966. The Maxvale exclosure 

 is situated 15 km. and the Lanherne exclosure 150 

 km. Avest of Charleville. Plant density and posi- 

 tion were recorded within an 80-cm. wide tran- 

 sect through the center of each exclosure. For 

 purposes of analysis each transect was divided 

 into 80 cm.-by-80 cm. quadrats. The exclosures 

 have been read irregularly since 1966. 



Changes in plant density under grazing were 

 determined by re-recording, in 1969, a 15-km. 

 length of a 2-m. wide belt transect laid down at 

 Humeburn (150 km. southwest of Charleville) 

 in 1965. The vegetation on this transect and the 

 method of recording it are given by Burrows and 

 Beale (3). For purposes of the present analysis, 

 the transect was divided into 200-m. 2 quadrats. 

 Those portions of the transect traversing alkaline 

 soils, on which E. gilesii rarely occurs, were 

 omitted. Stocking rate for the period 1965-69 

 was light and never exceeded one sheep to 5 

 hectares. 



