This paper examines the robustness of explanatory variables in cross-country economic growth regressions. It introduces and employs a novel approach, Bayesian Averaging of Classical Estimates (BACE), which constructs estimates by averaging OLS coefficients across models. The weights given to individual regressions have a Bayesian justification similar to the Schwarz model selection criterion. Of 67 explanatory variables we find 18 to be significantly and robustly partially correlated with long-term growth and another three variables to be marginally related. The strongest evidence is for the relative price of investment, primary school enrollment, and the initial level of real GDP per capita.

No abstract

Abstract. Pythiosis (204 cases, 77%), basidiobolomycosis (48 cases, 18%), and conidiobolomycosis (14 cases, 5%) were diagnosed morphologically from 266 horses with phycomycosis. All lesions were cutaneous ulcerative granulomas and three horses with pythiosis had metastatic lesions in regional lymph nodes. Lesions of pythiosis contained characteristic yellow, coral-like coagula and had a fibrotic surface containing sinus tract openings. Basidiobolomycosis was characterized by infrequent small yellow coagula and a yellow line of fungal invasion beneath an edematous surface. Lesions of conidiobolomycosis had numerous small coagula and a nasal location. There were minor differences in inflammatory cell populations within the granulomatous lesions. Most differences were associated with coagula size and fungal morphology. Coagula were composed of collagen, fibrin, cellular debris, degranulated and whole eosinophils and hyphae. Histochemistry revealed no major differences among the three diseases. Pythium sp hyphae were 2.6 to 6.4 pm in diameter, had thick walls, and occasionally were septate. Basidiobolus haptosporus hyphae were 5.1 to 20.5 pm in diameter, had thin walls, and commonly were septate. Conidiobolus coronatus hyphae were 5.1 to 12.8 pm in diameter, had thin walls, and commonly were septate. A perihyphal eosinophilic cuff (Splendore-Hoeppli phenomenon) with a radius of up to 20 pm was associated with the latter two fungi. Ultrastructurally, Pythium sp was composed of a thick, single density cell wall while B. haptosporus and C. coronatus had thin, double-layered cell walls.

Although 'phycomycosis' is a common disease of horses in northern Australia little is known about the causative fungi. In this paper the laboratory methods for diagnosis are described. These revealed 38 cases caused by Pythium sp. (Hyphomyces destruens), 6 cases caused by Basidiobolus haptosporus and 2 caused by Conidiobolus coronatus. Laboratory studies on the chemotatic behaviour of zoospores of Pythium sp. showed that they were strongly attracted to both animal hairs and plant tissue. Because of this behaviour a simple baiting method using human hair was used to trap the fungus from water samples taken from different locations near Townsville, Queensland. B. haptosporus was isolated by indirect culture from the faeces of 5 bearded dragon lizards (Amphibolurus barbatus) captured near a small creek near Brisbane. By combining previously published information with data obtained in the present study, ecological life-cycles for Pythium sp., B. haptosporus and C. coronatus are proposed. They include mechanisms whereby horses may become infected.