Content uploaded by Richard Primack
Author content
All content in this area was uploaded by Richard Primack on Mar 06, 2016
Content may be subject to copyright.
A preview of the PDF is not available
Insect pollination in the New Zealand mountain flora
Abstract
A wide variety of lepidopterans, bees, flies, and beetles visit the flowers of most species of New Zealand montane plants. Of the 82 plant Species which were well-collected, 4 species (Pterostylis sp., Pratia macrodon, Mazus radicans, and Dracophyllum acerosum) exhibited specialised pollination relationships with an insect order; 3 species (Cyathodes fraserii, Microris unifolia, and Thelymitra venosa) are not apparently visited by insects; and the remaining 7S species are visited by a variety of insects in 2 or more orders. Introduced plant species in the Composite family are visited predominantly by introduced bumblebees. Bees in the genera Lasioglossum and Leioproctus are abundant flower visitors. The most common lepidopteran flower visitors are in the families Noctuidae, Geometridae, and Pyralidae, and the genus Lycaena. Dipterans, in particular tachinids and syrphids, are numerically the most abundant flower visitors, and visit a wide range of species. The syrphid Melangyna novaezelandiae visits the flowers of more plant species than any other flower visitor. Beetles are typically not abundant, and do not move often between flowers. Species in the genus Hebe might be expected to have different insect pollinators from species in the Composite family. because of the quite different floral characteristics in these groups. However, there is no general difference in the insects visiting the flowers in the genus Hebe, the family Compositae. and the remaining species. indicating a lack of specialisation for particular pollinators. Individual flowers of most species last more than 4 days, so that several consecutive days of bad weather need not prevent a flower from being pollinated. There is no obvious relationship between the biogeographical Origin of plant species and the types of insects visiting its flowers.
... Alpine ecosystems in New Zealand are characterized by a lack of social bees, where most of the alpine plants are visited by dipteran insects and solitary bees (Newstrome and Robertson 2005). Small-sized owers with white or yellow colors are common in New Zealand alpine, and this trend has been thought to be related to the dominance of y pollinators, i.e., y-pollination syndrome (Wardle 1978;Primack 1983). Analysis of owering patterns of New Zealand alpine-plant communities is extremely important to clarify the relationship between y pollinators and owering phenology, but phenological studies on alpine-plant communities are scarce in New Zealand. ...
... The present study demonstrates that syrphid and non-syrphid ies are dominant pollinators of alpine plants in New Zealand, consistent with the ndings in previous studies (Primack 1983;Newstrome and Robertson 2005). Although beetles (leaf beetles and weevils) are accounted for about 10% of ower visitors, they are largely pollen or ower feeders and seemed not to function as pollinators (Primack 1983 The composition of ower visitors in the Mongolian grassland was very different from East Asian alpine ecosystems, where diverse insect groups frequently visited owers: non-eusocial bees were common, butter ies were major ower visitors, and syrphid ies were not a major group among dipteran insects. ...
... The present study demonstrates that syrphid and non-syrphid ies are dominant pollinators of alpine plants in New Zealand, consistent with the ndings in previous studies (Primack 1983;Newstrome and Robertson 2005). Although beetles (leaf beetles and weevils) are accounted for about 10% of ower visitors, they are largely pollen or ower feeders and seemed not to function as pollinators (Primack 1983 The composition of ower visitors in the Mongolian grassland was very different from East Asian alpine ecosystems, where diverse insect groups frequently visited owers: non-eusocial bees were common, butter ies were major ower visitors, and syrphid ies were not a major group among dipteran insects. The species diversity of bumble bees is especially high in inner East Asia including Mongolian grasslands (Naeem et al. 2019), but the frequency of other bees was much larger than that of bumble bees in our observation. ...
Flowering phenology of alpine-plant communities is determined by the interaction between abiotic and biological factors. Bees (especially bumble bees) and flies are major pollinators in alpine ecosystems. The abundance of bumble bees consistently increases with seasonal progress reflecting the colony development cycle, while the abundance of flies often fluctuates unpredictably. Responding to the seasonal dynamics of pollinators, flowering phenology of alpine-plant communities may also vary between bee-visited and fly-visited plants within and among regions. We compared the relationship between flower-visitor composition and flowering phenology across geographic regions: fly-dominated alpine in New Zealand, subtropical alpine in Taiwan, mid-latitudinal alpine in central and northern Japan, and high-elevation grassland in Mongolia. Thermal gradient was a fundamental factor regulating flowering patterns across regions, and clear seasonality at higher latitudes created diverse flowering patterns at a community scale. Flower production of fly-visited plants was less predictable with large variation, whereas that of bee-visited plants showed consistent patterns across regions reflecting the seasonality of bees. In New Zealand, most plant species were linked to syrphid and/or non-syrphid flies, but the network structure between insects and plants varied between sites. The network structures of the East Asian alpines were commonly constituted by syrphid flies, non-syrphid flies, and bumble bees, and these groups had specific niche width. In the Mongolian grassland, many insect groups formed diverse networks with small niche overlap. These results suggest that the flowering phenology of alpine-plant communities is influenced by the seasonal activity of bee pollinators under the climatic restriction in each region.
... This asymmetry was postulated as facilitated by the East Asian monsoon, hybrid incompatibility, and wind-mediated pollen dispersal. Species of Gaultheria can be pollinated by insects and wind [58,59]. In our study, east-to-west gene flow of the G. crenulata group may be enhanced by pollen-mediated movement influenced by the East Asian monsoon system, the significant role of which in shaping population genetic structure has also been demonstrated in plants such as Primulina Hance [60] and Quercus chenii Nakai [61]. ...
... As a secondary calibration point, we used the stem node age of the G. crenulata group [100] with a priori mean = 4.06 Ma, SD = 1 Ma, and 95% CI = 1.7-7. 59 ...
Background
The influence of native secondary succession associated with anthropogenic disturbance on the biodiversity of the forests in subtropical China remains uncertain. In particular, the evolutionary response of small understory shrubs, particularly pioneer species inhabiting continuously disturbed habitats, to topographic heterogeneity and climate change is poorly understood. This study aimed to address this knowledge gap by focusing on the Gaultheria crenulata group, a clade of small pioneer shrubs in subtropical China.
Results
We examined the genetic structure and demographic history of all five species of the G. crenulata group with two maternally inherited chloroplast DNA (cpDNA) fragments and two biparentally inherited low-copy nuclear genes (LCG) over 89 natural populations. We found that the genetic differentiation of this group was influenced by the geomorphological boundary between different regions of China in association with Quaternary climatic events. Despite low overall genetic diversity, we observed an isolation-by-distance (IBD) pattern at a regional scale, rather than isolation-by-environment (IBE), which was attributed to ongoing human disturbance in the region.
Conclusion
Our findings suggest that the genetic structure of the G. crenulata group reflects the interplay of geological topography, historical climates, and anthropogenic disturbance during the Pliocene–Pleistocene-Holocene periods in subtropical China. The observed IBD pattern, particularly prominent in western China, highlights the role of limited dispersal and gene flow, possibly influenced by physical barriers or decreased connectivity over geographic distance. Furthermore, the east-to-west trend of gene flow, potentially facilitated by the East Asian monsoon system, underscores the complex interplay of biotic and abiotic factors shaping the genetic dynamics of pioneer species in subtropical China’s secondary forests. These findings can be used to assess the impact of environmental changes on the adaptation and persistence of biodiversity in subtropical forest ecosystems.
... Muscidae are an abundant and speciose element of montane pollinating fauna (Primack 1983;Inouye & Pyke 1988) and one of many flies which are generalist in their flower visiting (Lord 2008). Historically, Bibionidae and Sciaridae emerge as a minor element of many pollinating faunas (Armstrong 1978;Primack 1983;Inouye & Pyke 1988). ...
... Muscidae are an abundant and speciose element of montane pollinating fauna (Primack 1983;Inouye & Pyke 1988) and one of many flies which are generalist in their flower visiting (Lord 2008). Historically, Bibionidae and Sciaridae emerge as a minor element of many pollinating faunas (Armstrong 1978;Primack 1983;Inouye & Pyke 1988). Recently, sciarids were shown to be frequent and dominant pollinators in moist forest understorey and sub-alpine meadow, particularly of plants with small and unspecialised flowers (Mochizuki & Kawakita 2018); there are several similarly configured plants common in the burned and heath quadrats. ...
In a world where fire is becoming more frequent and extensive, there is an opportunity and an obligation to understand the effects of burning on historically fire-free ecosystems. The nature and duration of fire effects on invertebrate species and assemblages in these ecosystems are particularly poorly understood. We compared the invertebrate assemblages from plots burned 54 years prior to sampling with neighbouring ancient sub-alpine coniferous and deciduous scrub on Mt Field, Tasmania, Australia. We tested the hypotheses that structural and floristic changes in the vegetation resulting from the fire influenced invertebrate assemblages and taxon distributions. We also tested whether capture through vacuuming was more effective in determining persistent differences resulting from fire than two aerial capture techniques. The legacies of fire on structure and flower availability affected the distributions of several invertebrate taxa. The taxon composition of invertebrate communities was most strongly differentiated between burned and unburned samples within the vacuum samples, with lesser effects from the aerial sampling techniques. Thus, the impact of fire on invertebrates persists for many decades, even when sampling areas are close to fire boundaries. The importance of preventing any burning of such fire-sensitive vegetation is reinforced by the significant reduction of several associated invertebrate taxa.
... Before humans arrived in Aotearoa New Zealand (NZ), forests covered 85 percent of the land [1] with an overabundance of small native white or pale flowers, where rewards such as nectar and pollen are freely available to any insect pollinator or floral visitor [2]. These floral traits are commonly observed in plants with a generalist pollination syndrome known as "small bee syndrome" which have corresponding short tongue length [3]. ...
Historic pollination networks are important to understand interactions between different plant and pollinator species, as well as to differentiate between causes and consequences of present insect population decline. Natural history collections in museums store biological proxy data, which is used to reconstruct historic pollination networks of bumble bees. Four bumble bee species ( Bombus terrestris , B . ruderatus , B . hortorum and B . subterraneus ) were introduced to Aotearoa New Zealand in 1885 specifically for pollination purposes. Pollen samples were collected from museum specimens of three of the four NZ species of bumble bee (excluding B . subterraneus ) collected between 1954 and 1972 from 56 locations across the South Island, New Zealand. The most common plants identified on all three bumble bee species were Calluna vulgaris (heather), Ulex (gorse), Cytisus (broom), and Trifolium repens (white clover). However, all three bumble bee species also carried pollen from several native plants (e.g. Arthropodium , Weinmannia , Plagianthus , Quintinia , Veronica , Melicytus ) and potentially had been involved in the pollination of these species. This study adds new plant species known to be foraged upon by bumble bees in Aotearoa New Zealand. Further studies on pollination networks in New Zealand will help us understand any changes in host plant preferences over time and after the time period covered by this study.
... The alpine biota includes a diverse assemblage of pollinators. These include native bees, Lepidoptera (Noctuidae, Geometridae, and Crambidae), and syrphid flies (Primack 1983;Campbell et al. 2010) and may include many day-active Coleoptera, especially Scirtidae, aleocharine staphylinids, Melyridae, and flower and broad nosed weevils (Barratt and Kuschel 1996;Klimaszewski and Watt 1997;Minoshima et al. 2018). Often, flower visitation isn't enough to confirm pollination directly, and gut contents need to be examined as well as follow-up studies that confirm successful pollen transfer on host plants (Minoshima et al. 2018). ...
New Zealand alpine environments host a diverse assemblage of insect lineages, with virtually every major insect group represented. The modern mountain ranges of New Zealand are relatively young and large areas of habitat above the tree line have only been in continual existence for the past one million years. We discuss the geological history and physical characteristics of New Zealand alpine environments and the resulting selective pressures placed on insect species. Some notable alpine taxa and previous faunistic research is highlighted. We discuss examples where single lineages have colonised the alpine zone and contrast these with larger radiations of alpine species which in some cases are the result of multiple colonisation events. The age of most alpine lineages is consistent with the young geological age of the mountains, nevertheless there are some much older alpine lineages of uncertain evolutionary history. We show that alpine species have employed a very broad range of morphological, physiological, and behavioural adaptations to survive in the alpine zone, and new studies are starting to unpick their genomic basis. Finally, we look to the future and assess threats to the unique New Zealand alpine insect fauna.
... We used the precision swab method on five families of insects that were collected while contacting the reproductive parts of flowers in New Zealand in the late 1970s (Primack, 1983). Families included: Syrphidae, Cerambycidae, Halictidae, Chrysomelidae and Mordellidae (two families featured: Syrphidae and Cerambycidae, Figure 3). ...
Historical datasets can establish a critical baseline of plant–animal interactions for understanding contemporary interactions in the context of global change. Pollen is often incidentally preserved on animals in natural history collections. Techniques for removing pollen from insects have largely been developed for fresh insect specimens or historical specimens with large amounts of pollen on specialized structures. However, many key pollinating insects do not have these specialized structures and thus, there is a need for a method to extract pollen from these small and fragile insects.
Here, we propose a precision glycerine jelly swab tool to allow for the precise removal of pollen from old, small and fragile insect specimens. We use this tool to remove pollen from five families of insects collected in the late 1970s. Additionally, we compare our method with four previously published techniques for removing pollen from pinned contemporary specimens.
We show the functionality of the precision glycerine jelly swab for removing small quantities of pollen across insect families. We found that across the five methods, all removed pollen; yet, it was clear that some are better suited for fragile specimens. In particular, the traditional glycerine jelly swab and the precision glycerine jelly swabs both performed well for removing pollen from bee faces. The shaking wash resulted in specimen fracture and residue left behind, the ethanol rinses left setae matted, and the glycerol swabbing left residue on the specimen. Additionally, we present photographs documenting the effects of these methods on pinned honey bee specimens.
The precision glycerine jelly swab opens up opportunities to sample pollen from a variety of insects in natural history collections. These pollen samples can be incorporated into downstream analyses for pollen identification either via microscopy or DNA sequencing, and the resulting plant–insect interaction data can establish historical baselines for contemporary comparison. Beyond our application of this method to pollen on insects, this precision glycerine jelly swab tool could be used to explore pollen placement specialization or to sample bryophyte, fungal and tree fern spores dispersing on animals.
... Lasioglossum spp. are often generalist flower visitors to a wide range of native and exotic plant species throughout their biogeographical ranges (Donovan, 2007;Elliott et al., 2021;Heine, 1938;Kleijn et al., 2015;Neave et al., 2020;Primack, 1983;Shebl, 2012;Soper & Beggs, 2013;Thompson & Merg, 2008;Webb, 1994). They are also noted within urban environments (Bennet et al., 2018;Frankie et al., 2009;Soper & Beggs, 2013). ...
Many ground-nesting bees are known contributors to crop pollination, but only a few specialist pollinating species are managed within agricultural systems. On the whole, most species (including the very large genus Lasioglossum) have been ignored for their crop pollination contribution, largely because of their perceived inability to be effective pollinators. Lasioglossum species are globally very widespread, including across many different environmental and ecological niches. Scattered reports provide credible evidence that these commonly small bees are effective pollinators across a wide range of crops, particularly as they often nest in close aggregations and may be present in crops in large numbers. Given the frequency and flexibility of this genus as flower visitors of many crops, further research to assess their effectiveness as pollinators deserves much greater focus. This is starting to change as assessments begin highlighting their effectiveness as pollinators of a range of crops. Preliminary attention is now being placed on developing potential management (or at least population-influencing) practices to enhance their use.
Flowering phenology of alpine plant communities is determined by the interaction between abiotic and biological factors. Bees and flies are major pollinators in alpine ecosystems. The abundance of bumble bees consistently increases with seasonal progress reflecting the colony development cycle, while fly abundance fluctuates unpredictably. Responding to the seasonal dynamics of pollinators, flowering phenology of alpine communities is expected to vary between bee-visited and fly-visited plants within and among regions. We compared the relationship between flower-visitor composition and flowering phenology across geographic regions: fly-dominated New Zealand alpine, subtropical Taiwan alpine, mid-latitudinal alpines in central and northern Japan, and high-elevation Mongolian grassland. Thermal gradient was a fundamental factor regulating flower patterns across regions, and clear seasonality at higher latitudes created diverse flower patterns within communities. Floral abundance of fly-visited plants was less predictable with large variation, whereas that of bee-visited plants showed consistent patterns across regions reflecting the seasonality of bee activity. In New Zealand, most plants were linked to syrphid and/or non-syrphid flies. The network structures of the East Asian alpines were commonly constituted by syrphid flies, non-syrphid flies, and bumble bees, and these groups had specific niche width. In the Mongolian grassland, many insect groups formed diverse networks with small niche overlap. Overall, bumble bees are suggested to be a driver of diverse flowering phenology in alpine ecosystems. In contrast, flies may not be a powerful driver of flowering phenology. Pollination networks between bumble bees and alpine plants are expected to be sensitive to climate change.
Ecological networks describe the interactions between different species, informing us of how they rely on one another for food, pollination and survival. If a species in an ecosystem is under threat of extinction, it can affect other species in the system and possibly result in their secondary extinction as well. Understanding how (primary) extinctions cause secondary extinctions on ecological networks has been considered previously using computational methods. However, these methods do not provide an explanation for the properties which make ecological networks robust, and can be computationally expensive. We develop a new analytic model for predicting secondary extinctions which requires no non-deterministic computational simulation. Our model can predict secondary extinctions when primary extinctions occur at random or due to some targeting based on the number of links per species or risk of extinction, and can be applied to an ecological network of any number of layers. Using our model, we consider how false positives and negatives in network data affect predictions for network robustness. We have also extended the model to predict scenarios in which secondary extinctions occur once species lose a certain percentage of interaction strength, and to model the loss of interactions as opposed to just species extinction. From our model, it is possible to derive new analytic results such as how ecological networks are most robust when secondary species degree variance is minimised. Additionally, we show that both specialisation and generalisation in distribution of interaction strength can be advantageous for network robustness, depending upon the extinction scenario being considered.
Pollination mechanisms and pollinators are reported for a total of 137 species (75% of the non‐abiotically pollinated flora) as they occur at three altitudinal levels (subandean scrub: 2,200–2,600 m; cushion‐plant zone: 2,700–3,100 m; subnival feldfield: 3,200–3,600 m) in the Andean (alpine) zone on the Cordon del Cepo (33°17'S) in central Chile as part of community oriented research in reproductive biology in the high temperate Andes of South America. Only around 4% of the species studied failed to be visited by potential pollinators. Hymenopterans (principally bees) are important pollinators of 50% of the biotically pollinated flora, butterflies of 24% and flies of 46%. Other vectors include beetles, moths, and hummingbirds. An estimated 17% of the flora is anemophilous.
Bee species‐richness, specialist feeding, and melittophily reach maxima in the subandean scrub; thereafter, bees diminish rapidly in number, with bees pollinating only 13% of the subnival flora as contrasted with 68% of the subandean flora. Although fly and butterfly species‐richness also decline with increasing altitude, the proportions of species pollinated by these vectors actually increases. High‐altitude populations of melittophilous species with broad altitudinal ranges are invariably serviced by fewer bee species as compared with lower populations.
The rich bee fauna at the lower end of the Andean zone in central Chile appears to have resulted from upward colonization from that of the subtending lowland Mediterranean sclerophyllous woodland vegetation. Altitudinal variation in pollination spectra is discussed in relation to contrasting life history characteristics and different modes of thermoregulation in the insect groups involved.
Insect-host plant associations in New Zealand are characterised by a high degree (over 60% in some families) of host specificity, by a low incidence of monophages on the warmth-adapted plants, and a higher incidence on the cold-adapted plants. Fossil age and assumed provenance show no correlation with incidence of monophages on a plant genus. Many host associations agree with other Austral (as against Holarctic) situations, but many have arisen de novo, indicating the long isolation of New Zealand since the Cretaceous. The scanty information on pollinators suggests that pollinator-flower relationships are, by and large, non-specific.
(1) Variation in flowering time of individuals in one population of each of three species of shrub was recorded over two growing seasons in montane scrub-grassland in the South Island of New Zealand. (2) There was considerable variation in flowering time within each population, but the flowering rank-order of individuals in different years was positively correlated. (3) Variation in flowering time was poorly correlated with the duration of flowering and the number of flowers and fruits per plant, except that variation in flowering time in Discaria toumatou (Rhamnaceae) was weakly positively correlated with the percentage fruit set in 1976-77. If both earlier and later flowering plants showed reduced fruit set this would suggest stabilizing selection, but there is no indication of such a pattern. Weak and inconsistent directional phenotypic selection for flowering time can be demonstrated for these two species however. (4) In the warm, dry summer of 1977-78, Leptospermum scoparium (Myrtaceae) and Dracophyllum spp. (Epacridaceae) flowered on average 9 days and 5 days earlier respectively and for 17 and 8 days shorter duration than in the cool, damp summer of 1976-77. Further, L. scoparium plants had a lower production of flowers and fruits in the second season in comparison with the first season. Plants of Discaria toumatou also flowered earlier in 1977-78, but the duration of flowering and flower and fruit production was greater in 1977-78 than in the 1976-77 season. (5) Patterns of variation in flowering time are also apparent among adjacent populations depending on altitude and on the major geographical units of the range of species. Variation in flowering time both at the individual and the population level may be an important adaptation by which selection and physiological mechanisms increase reproductive success.
Research on flower biology began in New Zealand in the early 1870s under the influence of Darwin's work on orchids, but from the turn of the century there was a decline in interest until the 1950s. Spring and summer are the main flowering periods, but many species flower in winter and examples are described. Of some 1800 indigenous species of flowering plants 12–13% are dioecious, c. 2% gynodioecious, and 9% monoecious. But in many cases the unisexuality is characteristic of a widespread family or genus and cannot be claimed as having evolved in New Zealand. The morphological differentiation between male and female flowers settles down at a level characteristic of the genus and the degree of differentiation need not reflect the time since differentiation began. In hermaphrodite species heterostyly is not known, and demonstrated examples of self-sterility are few. A classification of 649 species with attractive flowers gives 60.6% white, 17.2% yellow, 12.4% blue lilac or dark purple, 5.7% red to crimson, and 4% green. A large sample from the British Isles has 25.1% white flowers. It is emphasised the flowers classified as white are rarely completely so, that a white plus yellow group is important, and that not all the flower colours need have evolved under New Zealand conditions. Nectar and honey-dew from native plants provide useful honey sources, but work on nectar has been confined to cases of bee poisoning. Available pollinators are birds (7 spp.), bats (1 sp.), butterflies (16 spp.), and solitary bees (c. 40 spp.) as well as many species of moths, beetles, and flies and several introduced bees. The general impression is of widespread self-fertility in hermaphrodite plants and variety in respect of insect visitors. It is emphasised that although much attention, everywhere, has been given to methods of pollination, more attention should be given to the results, i.e., the percentage of ovules which produce seeds. And it is also emphasised that a better understanding of the characteristics of New Zealand flowers will be obtained by studying their relatives in other lands.
The history of “Cockaynes” in Sheffield is first outlined as well as aspects of Leonard Cockayne's early life, his family, and the sources of his private income.Cockayne's views on the evolution of divaricatingjuvenile forms and shrubs are summarised, and then examined with special reference to the three New Zealand species of Sophora. It is concluded that Sophora microphylla with ajuvenile divaricating form did not give rise by neoteny to the divaricating shrub S. prostrata, as Cockayne suggested, but that S. microphylla is derived from hybridisation between S. prostrata and S. tetraptera. It is also suggested that all species with divaricatingjuvenile forms arose by hybridisation between a divaricating shrub and a tree.The reasons why Cockayne favoured the inheritance of acquired characters are listed, and a restatement of the problems attempted in more modem terms. The selection pressures which could have led to divarication are also discussed, and it is considered that whereas some species may have evolved in drought-prone forests, others, such as Sophora prostrata, did not.The distribution and frequency of normal and cut-leaved juvenile forms in native aralioids are described, and the cut-leaved juvenile of Schefflera digitata shown to be genetically controlled and not environmentally induced as thought by Cockayne.Extracts are given from Cockayne's letters to Sir David Prain ofKew between 1912 and 1921, showing the development of Cockayne's ideas for a book on plant evolution in New Zealand. The unpublished and incomplete manuscript, now held at the Auckland War Memorial Museum, is described.The views of Lotsy on hybridism as a factor in evolution are then examined in relation to Cockayne's later work.
Although some of the plants of alpine and subalpine habitats in New Zealand are derived from ancestors that have been present since Paleogene, or possibly Cretaceous, times, many entered via Australia following the late Pliocene and subsequent elevation of mountains in Malaysia, New Guinea, Australia, and New Zealand. The New Zealand alpine and subalpine habitats only developed in the past few million years, during which they have been profoundly and repeatedly altered by Pleistocene glaciations and finally by man. As these habitats developed in isolation they were colonised mainly by plants derived from Australia, as a result of long-distance dispersal. Westward dispersal, in the face of the prevailing westerlies, is almost ruled out for Australasian plants. In New Zealand, many of them evolved rapidly to produce numerous species in consequence of the processes of adaptive radiation, interspecific hybridisation and recombination, and in many instances self-pollination. These large clusters of species, when compared with the relatively few members of the same groups in Australia, have previously led to the impression that the Australian representatives were derived from New Zealand. In fact, it seems to have been the unusual opportunities for rapid evolution, in response to a novel, isolated, and rapidly changing environment, that has led to the far greater representation of species in New Zealand.
Autumn initiation and successful overwintering of floral primordia are described for all 18 families and for 81 of the 100 alpine species examined. The high incidence of this phenomenon is not unexpected in view of its adaptive value and its widespread occurrence in alpine floras of the Northern Hemisphere. The causes of irregular flowering in alpine species of Aciphylla, Celmisia, and Chionochloa are discussed. A low-temperature requirement apparently prevents pre-winter opening of the fully developed buds in Caltha.
The period and intensity of flowering among spermatophytes was observed from spring 1963 to autumn 1966 at Cupola Basin, an alpine tributary of the Travers River, Nelson. Flowers were present from October to May, depending on conditions of weather and snow-cover in the three seasons, and among monocotyledons were observed to be restricted to December to March. Seventy percent of all flowering occurred between December and February. Onset of flowering was delayed by heavy snow, late spring thaw, and persistent rain in 1964–65, and in species typically distributed over a wide altitudinal range was restricted and more intense at higher altitudes. Onset of flowering was also later and briefer on southerly and other shaded slopes.