Introduction

Early biologists found it convenient to classify all living things as either animals or plants. To many people today, this grouping still seems perfectly adequate. However, examination of the life-forms that exist on Earth shows that this classification is unsatisfactory. Although there is a superficial resemblance between green plants and the fungi, these two groups are divided by profound biological differences. Unlike green plants, fungi cannot manufacture their own food from water and carbon dioxide by the process of photosynthesis. Rather, they require a supply of organic matter from which they can derive their energy. Fungal cellular composition is dissimilar from that of green plants, and the structural polymers of their cell walls are markedly different. Fungi are therefore now accorded their own status as a third kingdom. Furthermore, for many years the classification of microscopic organisms proved to be difficult. Photosynthetic microbes behave very differently from higher plants. It was therefore proposed towards the end of the nineteenth century that microscopic life forms should be classified as a fourth kingdom. This was the kingdom Protista, proposed in 1866, at a time when the scientific study of microbiology was in its infancy. This was, however, almost 200 years after Antonie van Leeuwenhoek described ‘animalcules’ following his development of the optical microscope.

During the twentieth century there have been many advances in microscopy, including the development of the electron microscope.

Isolation of bacteria

The earliest attempts to produce solid cultures included solidifying meat extracts with gelatin. Such media had two principal disadvantages: firstly, gelatin liquefies at about 37 °C, the optimum temperature for the growth of many human pathogens; secondly, many bacteria possess the ability to digest gelatin. Consequently, gelatin-based media tend to become liquefied under conditions where it is desirable to use a solid growth medium.

Agar is an inexpensive polysaccharide, obtained from certain seaweeds. In solution it can form a gel, and it provides an excellent substitute for gelatin as a solid support for microbiological media. Agar is generally resistant to microbial degradation and once gelled it remains solid at temperatures just below 100 °C. Once molten, agar suspensions remain liquid at temperatures of about 45 °C. This permits heat-labile supplements to be added to agar-based media without loss through heat degradation. Because of these properties, agar is the gelling agent used most widely in bacteriology.

During the earliest days of bacteriology, the most successful growth media were those derived from boiled extracts of meats of various types. Even now, brain–heart infusion broth is a rich growth medium often used to culture fastidious bacteria. Early culture media were very variable in their content. This caused problems with the standardisation of growth and also of bacterial characteristics. Today, many bacteriological growth media are still based upon peptones. These consist of a complex mixture of water-soluble products obtained from the hydrolysis of proteins derived from lean meats and other sources including heart muscle, casein and soya flour.

Introduction

Human life is sustained and enriched by the surrounding vegetation. The exploitation of the local vegetation is one aspect of human relationship with plants. The interactions between humans and plants may result in beneficial as well as deleterious effects.

Ethnobotany may be defined as the study of the biological bases of plant-human interactions and relationships at different levels of organization (e.g. ecosystem, community, individual), over geographic and social space, and along the evolutionary time scale. On one hand, the lifestyle and form of subsistence of inhabitants can be strongly influenced by the exploitation of local plants as well as by introduced elements (e.g. cultigens) and processes (e.g. livestocking). The diversity and availability of vegetal resources depends upon the local demand to make a living as well as the external demands for raw material. Intensification of certain extractive or ecologically non-compatible processes often destroys over time valuable and promissory resources. For the purpose of this chapter, I consider only the spontaneous plants which are exploited for local consumption rather than the cultivated exotics or the systems that modify the habitat for ecologically non-compatible production.

The tropical dry forest (tropical deciduous forest, ‘selva baja caducifolia’ or ‘bosque tropical caducifolio’ and associated vegetation types) extends along the length of Mexico, principally parallel to both the Gulf of Mexico and the Pacific Ocean coasts (see Murphy & Lugo, Chapter 2). It ranges from 0 to 1900 m in elevation and covers 157,800 km2 (or 8% of the Mexican territory) (Rzedowski, 1978). Flores et al. (1971), on the other hand, estimate that the Mexican tropical dry forest (‘selva baja caducifolia’) occupies 315,213 km2 (or 16% of the country).

Introduction

Temporal and spatial patterns of water availability are perhaps the most widespread determinants of the characteristics of tropical vegetation. Both locally and regionally, floristic diversity and composition, growth forms, phenology and demography are largely functions of hydrology. However, few studies have focused on how the reproduction of tropical forest plants changes along a gradient of water stress, except in regard to phenology, despite extensive field studies (especially of floral biology) and many reviews of the literature (e.g. Baker et al., 1983; Bawa & Hadley, 1990). In this review I emphasize comparisons among species and sites in dry forest and, especially, contrasts between drier and wetter localities, with respect to a series of topics. Do differences in water stress correspond to different patterns of mating and of evolutionary mechanisms, as determined by the frequency of sexual types and of self-incompatibility systems? How do the timing and duration of flowering and fruiting change across a gradient in the duration of water stress, and what determines such changes? What changes occur in the spectra of pollen and seed dispersal systems, and of available vectors? Lastly, extending our perspective to life histories raises questions of how reproductive effort changes due to water stress effects on mortality and growth.

This review is drawn from a heterogeneous literature. There are few comparisons of reproductive biology among localities differing in water stress, or in associated species, which focus on intraspecific or intrageneric variation. Thus many comparisons must be attempted on the basis of dissimilar species lists, and in spite of different methods. In a few cases some attention has been given to local differences between dry and moist sites.

Introduction

Holdridge (1947, 1967) developed a bioclimatic classification system by which the world's terrestrial biota may be categorized into approximately 120 life zones, each distinguished by climatic parameters that coincide with particular vegetational characteristics. Approximately 68 life zones are in the tropics and subtropics, of which 30 are dominated by forest of various types. Lugo, Schmidt & Brown (1981) estimated that 28 tropical and subtropical forested life zones are represented in Central America and the Caribbean, and 13 are found on the islands of the Caribbean. Despite this diversity, approximately half of the vegetation of Central America and the Caribbean is within the dry forest life zone (sensu Holdridge, 1967).

Dry forests are not infrequently referred to as deciduous forests, but the degree of deciduousness varies greatly (see below). Not all dry forests are conspicuously deciduous, and not all deciduous forests are dry forest. By Holdridge's criteria, tropical and subtropical dry forests are found in frost-free areas where mean annual biotemperature (a special calculation that reduces the effects of extreme temperatures) is above 17 °C, annual rainfall ranges from 250 to 2000 mm, and the ratio of potential evapotranspiration to precipitation is greater than one, to a maximum value of two. By these criteria, 49% (8.2 × 105 km2) of the vegetation of Central America and the Caribbean is considered dry forest (Brown & Lugo, 1980). Africa has the most dry forest (16.5 × 106 km2; 73% of the continent's vegetation); worldwide, about 42% of all intratropical vegetation is dry forest. Global patterns in dry forest distribution and overall ecological characteristics relative to wetter tropical and subtropical forest ecosystems were reviewed by Murphy & Lugo (1986a).

Introduction

Tropical dry forests constitute a large set of plant communities occurring under climates characterized by highly seasonal distribution of rainfall (Murphy & Lugo, 1986). The actual percentage of deciduous woody components varies from 100% to 40% depending on the specific forest type and its location within the rainfall gradient (Beard, 1955; Sarmiento, 1972; Hegner, 1979; Mateucci, 1987). In general it is assumed that along a gradient of rainfall in the lowland tropics, under similar temperature conditions, the proportion of deciduous woody components increases more or less linearly as the amount of rainfall received per year decreases below about 2000 mm (Walter, 1973). The structure of the community changes along these rainfall gradients in terms of community height, density of ground cover, proportion of trees and shrubs, and the occurrence of lianas, epiphytes and hemiparasites. The functional attributes of the structural components also change more or less monotonously along the gradients, thus the proportion of deciduous trees and shrubs tends to increase, the presence of epiphytes and hemiparasites is reduced, and at certain levels of the gradient the lianas (woody climbing plants) are important. Along the same gradients the proportion of succulent plants, including Crassulacean acid metabolism (CAM) performing cacti and stem succulent, drought deciduous trees, increases. A growth form showing dominance toward both extremes of the rainfall gradient is the evergreen type of woody plants. The evergreens at each extreme, however, may be separated according to their leaf structure and drought resistance. In this review I will discuss the differentiation of plant components in neotropical dry forests according to morphological and eco-physiological features.

Introduction

In Africa, the term dry forest covers vegetation types dominated by a more or less continuous tree cover (70%), experiencing pronounced drought during more than three months per year, and occurring within the savanna biome. They may be called (open) woodlands or (dense) dry forests according to tree density and understory structure (Menaut 1983). The Yangambi classification establishes the following (Boughey, 1957a, b; Monod, 1963; Aubréville, 1965).

• A woodland has an upper stratum of deciduous trees of small or medium size, with their crowns more or less touching above a sparse woody understory. Tree density is high enough to affect the herbaceous stratum which differs floristically from the adjacent savanna. The ground layer consists of grasses, herbs and suffrutescent plants in sufficient density to allow for annual burnings. The canopy of a woodland tends to be dominated by one or very few species.

• A dry forest, strictly speaking, is defined as a closed stand with several woody strata. The grass layer, when present, is weak and discontinuous, only allowing for episodic and sparse fires. In most cases, the trees of the upper stratum are deciduous whereas the understory is composed of evergreen and/or deciduous shrubs which differ from the canopy floristically. The canopy is multispecific and often devoid of woodland dominants. In both dry forest and woodland, the tree species which make up most of the canopy are present but never dominant in the surrounding savanna.

Some authors have considered dry forests to be a tropophilous extension of the rain forest, with adaptations to xeric conditions in characteristics of stems but not of leaves (Schnell, 1976–7).

Introduction

The tropical dry forests of Mesoamerica constitute an ecological theater where a plethora of herbivory plays occur with a great variety of forms, intensity and spatio-temporal variation. In contrast to other types of tropical forests, the herbivory events in dry forests are frequently quite apparent or visible to the eyes of the professional or even the casual observer. In these forests, for example, complete: defoliation events are not so uncommon, particularly on some deciduous (as opposed to evergreen) species of trees and shrubs, and these occurrences suggest that massive defoliations and insect ‘outbreaks’ are not exclusively extra-tropical phenomena. Likewise, levels of seed predation for some individuals in some species can sometimes be in the order of 100% (e.g. Acacia cornigera: R. Dirzo, personal observation).

From the point of view of the consumers, the protagonists of these ecological plays comprise two major groups.

1. 1 The array of vertebrates spans four orders of magnitude in body weight. In Santa Rosa National Park (‘SRNP’, Guanacaste Province, Costa Rica), these include forest understory mammals ranging from tapirs (Tapirus bairdii) (200,000 g) and white-tailed deer (Odocoileus virginianus), which actively consume seeds and foliage, to 30 g seed-eating mice such as Liomys pictus or Oryzomys palustris, and forest canopy mammals ranging from leaf- and flower-eating howler monkeys (Alouatta palliata) (8000 g) to small (40 g) leaf- and seed-eating rats such as Nyctomys sumichrastii.

2. 2 The array of invertebrate herbivores is extremely diverse. In SRNP these include not less than 110 species of seed-eating beetles (Janzen, 1980), and 3140 species of folivorous caterpillars (Janzen, 1988).

Likewise, from the point of view of the resource base, the plants also comprise a genetically diverse array.