ABSTRACT In Mediterranean‐climate shrublands, fire‐return intervals are decreasing, leading to negative impacts on native communities. Most research on this topic has been conducted on woody species, while little is known about the impacts of increased fire frequency on perennial herbaceous species. Geophytes are an ecologically important floral component, contributing to native diversity and often creating conspicuous postfire floral displays. Our long‐term observations of the chaparral geophyte Toxicoscordion fremontii provide information about its life history while offering insights into potential impacts of altered fire regimes. Over two decades and through two wildfires, we tracked individuals in two sites in Santa Barbara County starting in 1990, when half were burned and the other half unburned. Every year through 1999, we measured plant size and reproduction. After a large wildfire in 2009, when both sites burned, a majority of individuals were relocated and plant traits recorded. Most individuals survived the second fire and are long‐lived—lifespan of plants was at least 44 years and possibly over 85 years. Flowering was essentially restricted to the first spring postfire. Flower production was positively correlated to plant size, suggesting growth between fires is crucial. Fruit set did not increase with floral production, suggesting resource limitation. Although this species is fire‐resilient, more frequent burning could have significant negative consequences. Notably, with less time to accumulate carbohydrates available for postfire flowering, seed output and subsequent seedling establishment would be reduced. We recommend further study, including comparisons of postfire reproduction in populations with varying burn histories.

Obligate seeding plants—those that are killed by fire and whose persistence depends on soil-stored seedbanks—may be particularly susceptible to indirect effects of habitat loss and fragmentation. Especially for rare species, fragmentation can create conditions that reduce the likelihood of achieving sufficient seed bank stores due to changes in the interactions of the remaining plants with their pollinators or their natural enemies such as herbivores or seed predators. Arctostaphylos morroensis (Morro Manzanita) is an endemic obligate seeder whose distribution has been reduced greatly to a small portion of coastal California in habitat that is fragmented by development. We examined the reproductive ecology of A. morroensis to determine the factors that affect seed input to the soil seed bank. In stands of different ages and percent cover of Morro Manzanita, we observed insect pollinators, and measured flower and fruit production, and rates of fruit predation over two years, hypothesizing that, based on previous literature on obligate-seeders, fruit set should be relatively high across all stands, fruit set should be resource-limited, and seed input should be relatively high. Our observations suggest that A. morroensis reproduction is dependent on pollinators, which were primarily bees. Contrary to our predictions, we found that fruit set was relatively low (averaging 10–18% over both years) at all stands and appears to be pollinator-limited. Fruit predation rates were high, with the majority of fruits in experimental trays removed in a matter of weeks. We suspect that seed input in this rare species is strongly limited by low fruit set and high seed predation.

Morro manzanita Arctostaphyos morroensis (Ericaceae) is a long-lived, shrub endemic to San Luis Obispo County, southern California, USA. It was listed as threatened under the U.S. Endangered Species Act in 1994, with identified threats being residential and urban development, including lack of protection on private land and lack of management on public lands, competition with invasive non-native plants, and risks of extinction associated with small and isolated populations. Our goal in this paper is to summarize and supplement the current knowledge of Morro manzanita. We review the literature on the species' description, reproductive ecology, germination cues, short-term response to fire. and distribution. We conducted field surveys to report on long-term response to fire, resampling the previously studied prescribed burn site 25 yr post-fire. Finally, we summarize the current land management of sites that support Morro manzanita and threats faced by this species. We conclude with specific recommendations for management and future study towards supporting conservation of this species and its maritime chaparral community.

Abstract: Mode of reproduction in the genus Opuntia varies among species but often includes both vegetative and sexual reproduction, with the latter often facilitated by animal seed dispersal. In this multi-year study, we examined fruit and seed production and seed dispersal in the Mojave prickly pear (Opuntia phaeacantha) at two sites in the southwest Mojave Desert. Between 2015–2020 we counted fruits on randomly selected cactus patches, quantified fruit losses from those patches over time, collected fruits, and extracted and counted seeds. To assess seed dispersal, we collected fresh mammal pellets in belt transects at one site from 2016–2018; pellets were examined for Opuntia seeds. To assess animal fruit consumption and removal, we installed camera traps at one site. We baited selected patches with fruits and photographed animals that consumed or removed fruits. Fruit numbers varied widely among years and sites, although fruit production was not significantly correlated with climatic variables. Fruit losses were high at both sites, occurring more slowly in years of high fruit production. Seeds per fruit also varied with means ranging from 65 to 125 seeds. Rabbit and deer (Odocoileus hemionus) pellets were abundant at one site where we found 0.02 seeds per rabbit pellet but none in deer pellets. Camera traps baited with fruits revealed that they disappeared more quickly from patch edges than from patch interiors. Desert cottontails (Sylvilagus audubonii) and California jackrabbits (Lepus californicus) dominated photographs at patch edges while nearly all interior photographs were of white-tailed antelope squirrels (Ammospermophilus leucurus). In summary, although the number of seeds produced by O. phaeacantha is highly variable, the total number per site is high in some years, and fruits are consumed and seeds spread by animal dispersers.

On 1 September 2020, we lost a legend in the science of ecology: Dr. Joseph (Joe) Hurd Connell, who died aged 96 (Fig. 1). Joe’s research and conceptual writings have shaped the field since the publication of his highly novel PhD study about factors controlling the abundance and vertical distribution of two barnacle species on the intertidal seashore of Scotland. His pioneering field experiments and unmatched long-term monitoring studies transformed the field of community ecology. His example moved the discipline from a predominantly descriptive endeavor of cataloging and interpreting spatial and dynamic patterns in nature to an experimental, hypothesis-driven effort aimed at understanding mechanisms responsible for these observed patterns. He also wrote several highly synthetic review papers that refocused conceptual perspectives of the discipline, constructively challenged status quo paradigms, and identified important questions for future researchers. An insatiably curious, highly creative, warmly gregarious, and wickedly funny human being, Joe enriched and forever changed the lives of hundreds of friends and scientific colleagues, not to mention literally thousands of professional ecologists and their students who studied his publications or learned about his research from the pages of every ecology textbook and many introductory biology texts. In this essay, we share and celebrate the rich history of his life and his contributions to science and its practice. Some of the personal details we include come from the invited autobiographical accounts he wrote for Current Contents in recognition of five papers that had been designated “Citation Classics” (Connell 1981, 1987, 1988, 1989, 1992) and a personal profile he wrote for Peter Stiling’s textbook, “Ecology: Theories and Applications” (Stiling 2002, pp. 118–119). These essays reveal a lot about his motivations and personal circumstances. Margaret Connell, Joe’s spouse, generously provided additional details and helped us complete the timelines. Joe’s path to an extraordinary career in ecology was far from linear and included a rich assortment of life experiences. He was born the fifth of October 1923 in Gary, Indiana, USA, and attended a Catholic elementary school in Fort Wayne, Indiana, USA. Later, his family moved to Ellwood City, outside Pittsburg, Pennsylvania, USA, where his father was employed as an engineer in the steel manufacturing industry. Joe attended high school there. In fall 1941, he enrolled at Carnegie Institute of Technology, planning on becoming an engineer like his father. However, his life changed dramatically after the attack on Pearl Harbor and the United States’ entry into WWII in December 1941. In the following year, he joined the war effort and enlisted in the US Army Air Corps in December 1942. The Air Corps had a critical need for weather forecasters, so Joe was enrolled in a specialized training program in Meteorology at the University of Chicago (1943–1944). He was then stationed in the Azores as a commissioned officer (1944–1946) with the 1st Weather Reconnaissance Squadron, serving as meteorologist on weather surveillance flights across the North Atlantic in a modified B-25D Mitchell Bomber. He gathered data essential to the safe passage of American convoys crossing the Atlantic in support of the European Theater of the war. From an early age, Joe had enjoyed watching birds and identifying trees, but growing up in a small industrial town, where most professionals he knew were medical doctors, lawyers, or engineers, he did not recognize field biology as a viable career option. However, conversations with other army enlistees, who had pursued civilian careers in biology and wildlife management before joining the war effort, convinced him that a career as a field biologist was feasible. While in the Azores, Joe hiked around the islands observing birds and other wildlife. This was where his love of natural history became an enduring part of his makeup. After the war ended, Joe returned to the University of Chicago and completed his BSc in Meteorology in 1946, but now he was determined to pursue a career in field biology. Supported by funds from the GI Bill, Joe enrolled in a master’s program in Zoology at UC Berkeley in 1947 and earned his MSc degree in 1953 under the supervision of wildlife biologist Dr. Aldo Starker Leopold. For his master’s thesis, he set out to document the movements and home range of the brush rabbit, Sylvilagus bachmani, in a chaparral-grassland-dominated canyon just east of the Berkeley campus. Leopold recommended this study organism because he thought the rabbit was abundant and a potential game animal that had been little studied. As it turned out, this proved to be dubious advice. The rabbits were very difficult to capture; over the course of his 14-month study (March 1948 to May 1949), he sampled for 2251 trap nights, but caught only 40 rabbits, and several of these became “trap-happy,” returning 6-14 times to the same trap (Connell 1954). Although his findings were an important contribution to our knowledge of brush rabbit natural history at that time, Joe found the project “frustrating” and the results “pretty dull” (Connell 1981, Stiling 2002, p. 118). This discouraging experience may explain why, after collecting his rabbit trapping data in 1948–1949, it was not until June 1953 that Joe filed his master’s thesis. (By then, he had already started, in 1952, to collect data for his doctoral dissertation!) The serendipitous consequence of this tedious rabbit project was that Joe “vowed then to adopt a simple rule of thumb, namely, never again to study anything bigger than my thumb” (Connell 1981). This pledge and a fortuitous introduction to the little-known field experimental studies of the French marine ecologist, Harry Hatton (Hatton 1938) motivated Joe’s famous study of competitive interactions between two species of barnacles on the shores of Scotland, described below. Ironically, he later immersed himself in long-term studies of corals and rainforest trees, which spend only a small fraction of their lives at a size smaller than a human thumb. Unlike brush rabbits, however, they stay put and it is relatively easy to collect data from many of them! Taking a needed break from research after his master’s work, Joe tried his hand at secondary school teaching and taught biology for two years (1949–1951) at C. K. McClatchy Senior High School in Sacramento, California, USA. Teaching was hard work, but rewarding, and he might have continued in that profession had he not received notice that he had one year left of his GI Bill funds to use immediately or lose (Stiling 2002, p. 118). Two experiences he had while in graduate school at Berkeley were pivotal in determining his next career move (Connell 1992, Stiling 2002, p. 118–119). The first was a graduate seminar he had taken as a beginning student, in which he reviewed what he described as a “wonderful” paper published in 1947 by Edward Smith Deevey (Stiling 2002, p. 118). Deevey’s paper summarized and compared the limited number of life table datasets that had been collected from natural animal populations by that time. The most complete dataset included in the review came from Hatton’s study of settlement and survival rates in multiple populations of the intertidal barnacle, Balanus (= Semibalanus) balanoides, which lived on sheltered to exposed shores adjacent to St. Malo on France’s Brittany coast (Hatton 1938). Deevey was very impressed with Hatton’s results and specifically pointed out (p. 312) that B. balanoides “is a very favorable object for population research.” Joe took note that barnacles offered many advantages over brush rabbits for quantitative experimental studies of factors controlling the distribution and abundance of natural populations (and were smaller than his thumb!). The second event that shaped the trajectory of his career was his introduction to Dr. Charles Maurice Yonge, a sabbatical visitor to Berkeley from the University of Glasgow and a renowned specialist in the physiology and morphology of marine invertebrates, especially corals, mollusks, and crustaceans. With one year of GI Bill funding in his pocket and a clear vision of the kind of research he wanted to pursue, Joe moved back across the Atlantic to pursue a PhD in Zoology with Professor Yonge. For his dissertation research (1952–1955), Joe studied barnacle populations on the shores of the Isle of Cumbrae in the Firth of Clyde, Scotland, based at the Marine Station at Millport. Because his Scottish landlady, Mrs. Plant, charged him very modest room and board (only £9 per week), he was able to stretch his one year of GI Bill funding to three (Connell 1981). This is also where, in 1952, he met Margaret Harvey, a visiting graduate student researcher from Oxford University who was studying ctenophore biology. They married in 1954 in Exeter, England, Margaret’s hometown (Fig. 2). Wanting to better understand the details of Hatton’s (1938) study, Joe painstakingly translated from French to English the entirety of Hatton’s 107-page paper, discovering that not only had Hatton gathered detailed observational data on barnacle demographics, but had used controlled field experiments to examine the factors that shaped patterns of post-settlement survival. This was a highly novel approach, perhaps the first time that such experiments had been conducted in the field under natural conditions. Prior to that time, ecological experiments had largely been relegated to the laboratory environment. Hatton scraped clean patches of the rock surface and monitored larval recruitment and subsequent survival of B. balanoides. He similarly monitored co-occurring populations of the barnacle, Chthamalus stellatus, which lives higher on the shore than B. balanoides. Hatton primarily studied the effects of density and physical factors on survival, employing controlled transplant experiments to measure the effects on survival of tidal elevation, rock surface aspect with respect to sun exposure, and an individual’s age/size. He also performed surface wetting and shading manipulations to assess the effects of heat and desiccation. Deevey’s review paper and Hatton’s research greatly inspired Joe, who always gave credit where credit was due: “my career was shifted into a new direction by Hatton and Deevey, unbeknownst to them” (Connell 1992). Since Hatton had studied the effects of physical factors in controlling barnacle distributions across the tidal gradient, Joe decided to focus on the effects of biotic interactions on the same barnacle species, initially planning to investigate predation and intraspecific competition. Professor Yonge thought this an appropriate scope for Joe’s PhD dissertation, and cautioned him not to take on too much (Connell 1981). Joe, however, had taken a field ecology course at Oxford University, taught by Charles Elton, which convinced him that competition between species was an important biotic interaction structuring natural communities. Disregarding his advisor’s counsel, Joe surreptitiously added an experimental study of interspecific competition for space between B. balanoides and C. stellatus, species that were differentially distributed along the gradient of tidal height (Connell 1961a). Amazingly, this “side study” was not included in his dissertation, but turned into probably the most widely cited and influential study that Joe conducted. Joe surmised (Connell 1981) that this study received so much attention because there was a growing body of theory about interspecific competition, but little in the way of direct experimental tests demonstrating its influence on a natural animal population. Joe felt that the central study of his dissertation (Connell 1961b), while cited less frequently, was a more substantial and better paper. The two studies are highly complementary and employed controlled manipulations of barnacle densities, transplant experiments, and predator exclusion treatments to demonstrate the impacts of competition for space and predation on density and size structure. Adult C. stellatus are most abundant on the upper shore and rare below, even though their larvae recruit over a range of lower tidal levels. Conversely, B. balanoides adults are densest at mid-shore to low shore levels and rare at upper levels, despite their larvae settling over much of the tidal range. Joe’s experiments demonstrated that, while predation by the snail, Thais (=Nucella) lapillus, reduced the density of large B. balanoides, their preferred prey, the mortality they caused was not sufficient to prevent the faster-growing B. balanoides from competitively excluding C. stellatus from the mid-shore to low shore. When B. balanoides were manually removed from the mid-shore plots, C. stellatus survived and grew well, while suffering high mortality due to competition for space with B. balanoides in unmanipulated control plots. In fact, C. stellatus grew better at mid-low shore levels, where they were more frequently submerged and filter feeding, than on the upper shore, where they were exposed to the air for longer periods. B. balanoides cannot survive the desiccating conditions characteristic of the upper shore, so the more desiccation-tolerant C. stellatus occupies this spatial refuge, free of competitive pressure from B. balanoides, thereby ensuring coexistence of the two species on the same shore. Application of these innovative yet simple experimental techniques clearly revealed the mechanisms underlying the differential distribution of species along an environmental gradient. Joe’s approach inspired the use of controlled experimentation in hundreds, if not thousands, of similar studies across a wide variety of marine, freshwater, and terrestrial habitats. In a word, Joe’s dissertation study revolutionized the science of community ecology. It is no wonder his study’s results are featured in so many introductory ecology texts. In recognition of the special significance of the work, Joe was awarded the 1963 Ecological Society of America’s Mercer Award, given for “an outstanding ecological research paper published by a younger researcher.” Joe received his PhD in Zoology from the University of Glasgow in 1956. Following a postdoctoral study at Woods Hole Oceanographic Institute (1955–1956), where he studied the spatial distribution of two species of clams, he accepted a two-year instructorship in Zoology (1956−1958) at the University of California, Santa Barbara (UCSB), and was subsequently hired as an Assistant Professor of Zoology in 1958. He spent the rest of his career at UCSB, earning tenure to Associate Professor of Zoology in 1961, and a promotion to Full Professor of Zoology in 1966. He retired in 1991 as Professor Emeritus of Zoology and held the appointment of Research Professor of Biology from 1996 to the time of his death. When he returned to the west coast of the United States, Joe sought to test the generality of the patterns he had documented in Scotland. Soon after being appointed to the faculty at UCSB, Joe headed north to the Friday Harbor Marine Laboratories on San Juan Island off the northwest coast of Washington State, where a comparable assemblage of barnacles and their snail predators occupy the intertidal zone. Working at two sites on the shore of San Juan Island, Joe conducted a series of experiments, similar to those he had used in Scotland, to study the impact of competition for space among three species of barnacles. The interactions proved quite different in this site, where three species of predatory snails (Thais spp.) preyed on the barnacles, as opposed to the single species of predatory Thais in Scotland. Their combined impact strongly reduced the density of the most common barnacle species, Balanus glandula, so that competition for space among barnacle species on the low to mid-shore was insignificant (Connell 1970), a very different outcome than Joe observed in his dissertation research. This study was among the first to demonstrate that predation can mediate the intensity of interspecific competition and thereby promote the coexistence of potentially competing prey species. The site-dependent nature of the interactions that Joe documented in Scotland versus Washington State became a major conceptual theme (i.e., context-dependence) in community ecology 40 years later. After being awarded tenure in 1961, Joe decided he “would strike out in a new research direction—try something new or outrageous—without worrying about whether it might cause a hiatus in my publication record” (Connell 1987). After studying relatively low-diversity temperate marine intertidal ecosystems for more than a decade, Joe became increasingly curious about the mechanisms that maintain high levels of species diversity in tropical ecosystems. At that time, Joe subscribed to the prevailing theory that, over evolutionary time, the high productivity and purportedly stable climatic conditions of the tropics had selected for narrow, specialized niches and high levels of species packing along resource gradients. He recognized, however, that the assumptions and predictions of this model needed to be challenged with data. By collecting long-term, spatially explicit data on demographic dynamics, including patterns of recruitment, growth, and mortality, Joe aimed to evaluate the stability of these communities and discover the mechanisms that structure them. He chose to go “all-in,” successfully applying for a Guggenheim Fellowship and moving with his family to the tropics of Australia in 1962 to study the ecological processes structuring coral assemblages on the Great Barrier Reef. During that first year in Australia, Joe met John Geoffrey Tracey and Leonard Webb of the CSIRO Rainforest Ecology Unit, the only two Australian ecologists working in rainforest at that time (Connell 1987). Their discussions gave birth to the idea of a collaborative study of mechanisms maintaining species diversity in this iconic tropical habitat, and in 1963, Joe added rainforest ecology to his portfolio. These more species-rich communities do not lend themselves as readily to the small-scale experiments that Joe had so successfully used on rocky seashores to identify the mechanisms structuring those communities. Nevertheless, he and colleagues were able to conduct manipulations of avian and mammalian seed predators and litter disturbers at his forest site in North Queensland to assess their effects on seedling assemblages (Theimer et al. 2011). The major results of his tropical work, however, come from the analysis of long-term census records collected from permanent plots established on the Great Barrier Reef and in two rainforest sites in the state of Queensland. In 1962, during his first visit to Australia, Joe established replicate permanent 1-m2 plots, line-intercept transects, and belt transects in several different sub-habitats across the reef on Heron Island near the southern end of the Great Barrier Reef. Changes in the 1-m2 plots, including the recruitment of new colonies, were quantified from 35-mm color slides taken with an SLR camera positioned vertically over each plot during 36 visits in 26 of the 38 years between 1962 and 2000 (Fig. 3; Connell et al. 1997a, 2004, Tanner et al. 2009). A year later in 1963, Joe, working with Tracey and Webb, laid out two large permanent rainforest plots, one (1.7 ha) at Davies Creek in tropical North Queensland southwest of Cairns, and the other (1.9 ha) in subtropical South Queensland near O’Reilly’s Rainforest Resort within Lamington National Park just south of Brisbane (Connell et al. 1984, Connell and Green 2000, Green et al. 2014). Seedlings and small saplings were identified, measured, and mapped within several belt transects at each location, and larger trees were individually marked and censused throughout both plots. These plots have been re-censused by teams of field assistants every few years; surviving trees are remeasured, new recruits tagged, and deaths recorded (Figs. 4-6). Through the years, literally hundreds of graduate students and postdocs "groveled" on the rainforest floor with Joe, measuring seedlings and getting leech bites as part of a ritual for aspiring field biologists. In celebration of the 50th anniversary of the establishment of these plots, Harms and Green (2014) published a description of the history and contributions of what is now known as the Connell Plots Rainforest Network. These records of community structure and dynamics may be the longest ever collected from these habitat types. Joe immediately recognized that, to make meaningful headway in studying the mechanisms maintaining diversity in these hyper-diverse communities, he needed the expertise of skilled taxonomists to accurately distinguish the many morphologically similar species co-occurring in each habitat. In addition to being accomplished ecologists, Tracey and Webb were experts at identifying rainforest plants, including their juvenile stages. Drawing on Tracey’s extensive field observations and his help compiling large sets of forest survey data, Webb had published the first physiognomic-structural classification of Australian rainforest vegetation in the Journal of Ecology in 1959. On the coral side, Dr. Carden Wallace, of James Cook University and the Museum of Tropical Queensland, provided essential help identifying corals in his plots and belt transects. Several events raised questions in Joe’s mind about the generalization that tropical communities should be considered equilibrial or “balanced” systems, tightly regulated by density or frequency-dependent biotic interactions such as interspecific competition for limited resources, predation, and disease. The first was a teatime conversation Joe had with the renowned Australian population biologist, Charles Birch, during which Joe presented his working hypothesis of stable, tightly regulated tropical communities. Birch replied: “Fine, but also keep your eyes open for tropical weather up in Queensland—some of those cyclones might shake things up a bit” (Connell 1987). The second event that challenged his assumption was in fact the impact of a severe cyclone that passed over both Heron Island and one of his rainforest plots in 1967, causing considerable damage (Connell et al. 1997a, 2004). At that time, many researchers considered these episodes rare, annoying events that “wrecked my study,” rather than drivers of community dynamics whose effects might not be recognized without long-term monitoring studies. Joe subsequently broadened his view of the possible mechanisms that could maintain diversity in these systems to include disturbance. Joe’s detailed coral and rainforest studies yielded novel insights into the processes that maintain the extraordinary levels of species diversity in these two tropical ecosystems. The studies challenged the long-standing view that these were equilibrial assemblages of tightly co-evolved species. From his 38-year record of changes in the coral system (Connell et al. 1997a, 2004, Tanner et al. 2009), Joe documented the damage caused by multiple cyclones and subsequent patterns of recovery. Large storm waves associated with these storms battered the reef, breaking and displacing colonies, while shifting sediments harmfully abraded them. Partial or complete death and displacement of colonies opened space for new recruits; competitively dominant species were knocked back and recolonization by competitively inferior species enriched the coral assemblage. This episodic reset of coral assemblages to a younger, more diverse state by disturbances of intermediate frequency and intensity as a general mechanism by which diversity is maintained is what Joe called the “Intermediate Disturbance Hypothesis” (Connell 1978). In the Queensland rainforests, a different story emerged. When storms and other natural disturbances damaged or killed trees (Connell et al. 1997b), they opened light gaps that enhanced local recruitment and growth of juvenile trees. But other diversifying processes were also at work. Joe’s good sense to study all rooted life-cycle stages of his rainforest trees arose from his earlier research, which convinced him that both the supply side and what might be called the “sorting side” (i.e., the competition, predation, disease, and habitat filtering that can non-randomly sort among individuals) were important in community ecology. The youngest seedlings on his transects clearly recruited in spatial clumps, most often near the parent tree, and recruitment rates varied year to year (e.g., Connell and Green 2000). Even so, relatively quickly the pattern diversity of cohorts of seedlings increased as they matured into later-stage saplings. These observations suggested to Joe that a combination of limited seed dispersal and the frequency-dependent influence of natural enemies—especially those that are relatively host specific—could be important diversifying mechanisms for trees (and other sessile organisms). Joe published his idea in a book chapter (Connell 1971) soon after tropical biologist Dan Janzen (1970) independently published similar thoughts. Their ideas have been linked ever since as the “Janzen-Connell Hypothesis” (Hubbell 1980). The concept has had a resounding influence on tropical forest ecology and remains one of the key elements in our collective working model for tropical rainforest diversity and dynamics (Wright 2002). Joe’s empirical findings and conceptual writings forever changed the science of ecology. With a well-deserved nod to his little-known predecessor Harry Hatton, Joe deserves credit for introducing and popularizing the use of controlled, replicated experiments to investigate ecological hypotheses in nature. One measure of the impact of his work on the discipline is how heavily his papers have been referenced by other scientists: 11 papers or book chapters have been cited over 1,000 times each. To date, these 11 papers combined have been cited 32,136 times! Joe also made major contributions to the growth of ecological theory and concepts by writing highly synthetic and critical reviews that evaluated published research results pertinent to key ecological phenomena. Especially impactful reviews addressed the role of natural enemies in preventing competitive exclusion (Connell 1971), community interactions on rocky intertidal shores (Connell 1972), shifts in species interactions along environmental gradients (Connell 1975), mechanisms of ecological succession (Connell and Slatyer 1977), hypotheses explaining the maintenance of high diversity in tropical communities (Connell 1978), the prevalence and relative importance of interspecific competition (Connell 1983), and the evidence needed to judge ecological stability and persistence (Connell and Sousa 1983). Each of these reviews has been cited more than 1,000 times to date, with the succession paper receiving more than 5,200 citations and the tropical diversity paper receiving more than 11,600! Ironically, while Joe provided perhaps the most famous demonstration that interspecific competition can shape community structure and species distributions, he also actively challenged the notion that it was the overriding and ubiquitous factor structuring communities. Several of his reviews questioned this prevailing perspective (Connell 1980, 1983) and provoked those invested in the “competition” paradigm. In the arena of applied ecology, Joe served as inaugural Chair of the Marine Review Committee (1974–1990), which was charged by the California Coastal Commission with assessment of the environmental impacts of the San Onofre Nuclear Power Plant on the coast of southern California. Joe received many honors and awards, among them two Guggenheim Fellowships, the Ecological Society’s Mercer and Eminent Ecologist awards, Fellow of The American Academy of Arts and Sciences, and corresponding member of the Australian Academy of Sciences. Joe was insatiably curious about the world around him (natural and human), and he wanted to know about everything. Few, if any, ecologists have studied as many different organisms and natural communities as Joe did, including small mammals in chaparral, rocky intertidal invertebrates, soft-sediment clams, rainforest trees, corals, and desert shrubs. He thoroughly enjoyed opportunities to immerse himself in new natural history experiences and loved engaging with students over their research ideas and results, regardless of the study organism or habitat. He was unfailingly objective about his own ideas and data, adhering to the Popperian scientific method more closely than many scientists. He did not cling to pet hypotheses; if evidence falsified one of them, he gladly moved on to consider another. Ego did not drive his scientific judgments. Joe was also highly egalitarian in his interactions with students and colleagues, especially when it came to seeking critical comments on his manuscripts. He would relentlessly pepper colleagues and students with multiple drafts of every paper he wrote. He carefully considered every comment, no matter whether it came from an undergrad or a full professor. If something was unclear, it was his fault, not the reader’s. He worked hard to clearly communicate his results and conceptual ideas in his writing, Joe gave his graduate students a lot of freedom in their choice of study organisms and field sites. He mostly cared that the questions and hypotheses were clearly stated and testable and that the findings would make a novel and valuable contribution to our understanding of the natural world. Of his 18 PhD and 2 MA thesis students, 12 worked in marine rocky intertidal habitats, two in marine soft-sediment environments, two in subtidal kelp forests, one in chaparral shrubland, one studied salmon in aquatic habitats, one studied terrestrial snails, and one studied stream flatworms. Fifteen postdoctoral researchers worked with Joe. Most collaborated on data collection, data analysis, or manuscript writing associated with either his coral reef or rainforest study. Several were involved with assessment of the potential impacts of the San Onofre Nuclear Power Plant. Others pursued their own research projects. Joe considered them peers and their collaboration greatly enhanced and expedited the analysis and publication of some of his most important research results. Joe gave detailed feedback on his students’ work. When he finished working over their papers, the bath of red ink could be visually and psychologically jarring, but the feedback always resulted in a clearer exposition of ideas. Some of his most frequent corrections were marked “No NUAs” (nouns used as adjectives) or “EUW” (eliminate unnecessary words), mixed with a few good-natured curse words. Learning to accept constructive criticism was a key part of his regimen of scientific training. He was so committed to the value of scientific data that he actually wrote up and published the dissertation chapters of two of his PhD students who were unable to do so, assigning himself second authorship. Even though he spent long hours advising on research questions and study design, he never asked to be included as an author on his students’ dissertation publications, believing that his mentoring efforts were part of his job description and that sole authorship was in the best interest of his students’ careers. It was not only the professional success of his students and colleagues that he cared about; he was also attentive to their general well-being by offering thoughtful support and kindness during hard times and, on occasion, providing practical material support for fieldwork. For example, when one of his graduate students had no means of transport to off-campus field sites, Joe donated an old, but fully functional, family car to the cause. Joe was a very social person, whether it be a campfire “billy up” at lunchtime on the side of the dirt road to the Davies Creek site (Fig. 7; https://www.nma.gov.au/exhibitions/symbols-australia/billy), or a gathering at his home to celebrate a visitor or special event. He loved hearing what people had to say, and the conversations often moved to non-scientific topics like favorite movies and books. He loved Japanese samurai films, classical music, and Marx Brothers movies, often imitating Groucho’s witty mannerisms and clever remarks. And Joe was a huge fan of the LA Dodgers. He probably cheered them to a World Series victory shortly after his own final inning in 2020. Joe is survived by his wife Margaret, their four children, Andrew, Jane, Tim, and Kate, and his niece Mary Rollins Jones, along with 11 grandchildren and seven great grandchildren. We are deeply appreciative to Joe’s family for sharing him with us; it is such a precious gift. A graduate research fund has been established in Joe’s honor at UCSB: The Joseph H. Connell Field Ecology Research Fund. Donations may be made in two ways. Those who wish to donate online can use this link: https://giving.ucsb.edu/Funds/Give?id=78 (please add "for JH Connell Fund" in the notes.). For donations by check, please make it payable to the UCSB Foundation (with “for JH Connell Fund” in the memo line) and mail to: Office of Development #2013, Attn: Bethany Innocenti, UCSB, Santa Barbara, CA 93106-2013. Questions may be directed to bethanyinnocenti@ucsb.edu. We thank Margaret Connell for providing important details about Joe’s life. We are also grateful to Betsy Mitchell for comments that improved the manuscript.

Conservation of obligate-seeding shrubs from Mediterranean-climate regions is an international conservation priority. Morro manzanita (Arctostaphylos morroensis) is one such shrub whose persistence may depend on germination and establishment from soil-stored seeds following fire. However, fire has been virtually eliminated from its remaining habitat. Thus, conservation of A. morroensis may depend on actions that stimulate germination to establish new populations. We characterized seed banks in different-aged stands and examined viability and germination of A. morroensis seeds in response to various cues, including heat and charred wood. We found that seed density varied greatly among sites, although not increasing with stand age as predicted. Rather, the oldest stand had especially low seed densities and viability. Viability of A. morroensis seeds was low-on average similar to 4%-limiting germination. Surprisingly, similar to 40% of viable seeds germinated with no fire treatments. Neither heat nor charred wood alone enhanced germination; however, when combined the two resulted in highest germination. Seeds soaked in water prior to heat-and-charred-wood treatments had significantly reduced germination, suggesting that prescribed burns conducted in the wet season would result in a poor germination response. Characterization of the seed bank and determination of the cues stimulating germination can provide information vital to the maintenance of this and similar obligate-seeding species.

California's oak savannas and associated grasslands occupy roughly 3 million ha. They are defined by the presence of scattered oak trees, in monospecific or mixed stands, with an herbaceous ground cover, composed predominantly of annual grasses and forbs. Supporting natural communities rich with native flora and fauna, oak savannas and grasslands contribute significantly to the designation of the California Floristic Province as a "biodiversity hotspot". These systems also provide numerous benefits to humans: in the past sustaining indigenous cultures who depended on oak acorns for food, and in the more recent past and present as a source of wood, timber, livestock grazing, and recreation. Climate in California's oak savannas is Mediterranean, with cool, wet winters and hot, dry summers. Both climate and soil type exert broad control on the distribution of grasslands and co-occurring woody vegetation in California. At a finer scale, plant and animal interactions strongly influence community structure and dynamics, as do human impacts such as livestock grazing and the use of fire. Two of the most significant threats to the longterm sustainability of California's oak savannas and grasslands are land-use conversion to agriculture and residential development, compounded by low rates of natural regeneration in extant stands of oaks.

AbstractThe co‐existence of evergreen and deciduous oaks in Mediterranean‐climate savannas has motivated comparative studies on species’ physiological adaptations to light and drought, establishment niche differences in acorn production, dispersal and seedling herbivory, and differential sapling tolerance of ungulate browsing. Understanding how species’ differences collectively affect co‐occurrence or segregation requires long‐term studies that consider multiple life stages. We compare survival, height growth, and modeled population growth rate of the evergreen sclerophyll, Quercus agrifolia, to those of the winter‐deciduous, broadleaved Q. lobata in southern California savanna sites where the species naturally co‐occur. We evaluate species’ performances after 8.7 yr for four cohorts sown as acorns in 1997, 1998, 2000, and 2001 under four different levels of protection from cattle, deer, and rodents. Survival curves for the two species were closely similar, with large differences between cohorts associated with precipitation in the year acorns were sown. Survival and growth rates of Q. agrifolia seedlings and saplings were slightly lower, and were more reduced by ungulates and rodents, than those of Q. lobata. No Q. agrifolia plants (0/1431) exposed to ungulates (cattle and/or deer) and rodents survived to the end of the experiment. Of 1421 Q. lobata acorns planted in these treatments, only 19 plants (1.3%) survived 8.7 yr and none of these plants progressed out of the ungulate browse layer. However, when protected from both ungulates and rodents, survival rates increased significantly for both species, and Q. agrifolia increased height faster than Q. lobata. Simulated population growth rates increased slightly for plants protected from ungulates, and 95% confidence intervals spanned the stable population growth rate of 1. With exclusion of ungulates and rodents, sapling and small tree recruitment rates for both species far exceeded those needed to offset the rate of recent adult mortality, with modeled population growth rates of 1.07–1.11. Our results underscore the essential role of occasional high rainfall years for initial establishment, plus the key role of consumers in limiting early survival and growth of both evergreen and deciduous oaks in mixed oak savannas.

Herbivores alter plant biodiversity (species richness) in many of the world's ecosystems, but the magnitude and the direction of herbivore effects on biodiversity vary widely within and among ecosystems. One current theory predicts that herbivores enhance plant biodiversity at high productivity but have the opposite effect at low productivity. Yet, empirical support for the importance of site productivity as a mediator of these herbivore impacts is equivocal. Here, we synthesize data from 252 large-herbivore exclusion studies, spanning a 20-fold range in site productivity, to test an alternative hypothesis-that herbivore-induced changes in the competitive environment determine the response of plant biodiversity to herbivory irrespective of productivity. Under this hypothesis, when herbivores reduce the abundance (biomass, cover) of dominant species (for example, because the dominant plant is palatable), additional resources become available to support new species, thereby increasing biodiversity. By contrast, if herbivores promote high dominance by increasing the abundance of herbivory-resistant, unpalatable species, then resource availability for other species decreases reducing biodiversity. We show that herbivore-induced change in dominance, independent of site productivity or precipitation (a proxy for productivity), is the best predictor of herbivore effects on biodiversity in grassland and savannah sites. Given that most herbaceous ecosystems are dominated by one or a few species, altering the competitive environment via herbivores or by other means may be an effective strategy for conserving biodiversity in grasslands and savannahs globally.

Large grazing animals can have profound impacts on plant communities and soil properties; however, these impacts are not always uniform across or within regions. The distribution of features such as forage quality, water, or shade within a pasture can change the behavior of grazers and thus, the impact of their grazing. Where managed livestock grazing has been proposed as a conservation tool to enhance or maintain desirable plant communities, understanding how spatial variation between tree and intertree habitats within a savanna landscape affects the response of vegetation and soil properties to grazing will be critical for designing management plans for different sites. In this study, we used a previously established, long-term livestock grazing experiment in California oak [Quercus L.] savannas to investigate how the removal of grazing affected plant communities and soil characteristics underneath and outside of isolated tree canopies. In the oak understory, plant community composition shifted in response to livestock removal, largely due to a 68 − 400% increase in the relative cover of native species. Overall plant community composition in open grassland neighboring trees changed little in response to livestock grazing removal, yet we did see a decrease in species richness and diversity surrounding deciduous oaks as the dominance of the exotic annual Bromus diandrus Roth increased. The depth of plant litter increased 1 − 2 cm in both habitat types when livestock grazing was absent, along with minor changes in soil carbon, nitrogen, and bulk density. These results highlight how different habitat patches within savanna landscape can have varying responses to grazing removal and illustrate how challenging it will be to use grazing as a management tool to enhance the diversity of native species. In the oak understory, native species that are tolerant of herbivory may be absent or unable to coexist with non-native annual grasses. The abundance of understory habitat at a particular site may therefore be an important variable predicting the outcome of livestock grazing.

The relative importance of livestock grazing in limiting or enhancing oak recruitment remains unclear because results from previous studies have been contradictory. In Santa Barbara County, we have replicated large-scale planting experiments from 1997 to 2001 to determine the effects of cattle and other factors on seedling establishment of valley oak (Quercus lobata) and coast live oak (Q. agrifolia). We manipulated cattle grazing (ungrazed vs. winter-spring rotational grazing) and protection from small and large mammals. Rainfall and seed predation and herbivory by small mammals—most likely gophers and ground squirrels—significantly impacted rates of seedling recruitment. Exclusion of cattle alone has not significantly increased establishment of either valley or coast live oak. However, protected seedlings in pastures with cattle have higher survival and growth rates than protected seedlings in plots excluding cattle. Our results suggest that winter-spring livestock grazing can have indirect positive effects on oak establishment, by reducing herbaceous biomass and associated small mammals adjacent to protected seedlings. Contrary to expectations based on the relative abundance of natural oak seedling recruits, establishment and survival of coast live oak planted in our experiments has been significantly and consistently lower than that of valley oak.

The Santa Barbara County Oak Restoration Program was initiated in 1994 to determine the major factors limiting recruitment of valley oak (Quercus lobata) and coast live oak (Q. agrifolia). At Sedgwick Reserve in Santa Barbara County, California, we have replicated large-scale planting experiments in four different years to determine the effects of cattle and other ecological factors on oak seedling establishment in oak savannas and woodlands. In 33 large experimental plots (50 x 50 m) we planted acorns collected from Q. lobata and Q. agrifolia on the site. Fifteen of these large plots are controls, open to grazing, fifteen exclude cattle with the use of electric fence, and three are ungrazed in large ungrazed pastures. Within the plots, experimental treatments included: 1) protection from small mammals such as gophers and ground squirrels, 2) protection from large animals such as cattle, deer, and pigs, and 3) no protection from mammalian grazers. In winters 1997, 1998, 2000, and 2001, we planted approximately 1,000 acorns of each species. Results confirm that seed predation and herbivory by small mammals are a significant "bottleneck" to oak seedling recruitment on the landscape scale. Comparing results among years indicates that lack of late winter rainfall can significantly reduce oak emergence and establishment. Survivorship of protected acorns and seedlings is comparable in grazed and ungrazed areas.

ABSTRACT. Carnivore consumption of fruit is a principle means by which many fleshy-fruited plant species achieve long-distance seed dispersal. We examined carnivore dispersal of hollyleaf cherry (Prunus ilicifolia) seeds, specifically assessing the survival, desiccation sensitivity, and germination of seeds found in bear scats. Studies were conducted both in the laboratory and in 2 burn areas in Los Padres National Forest, California. Bear scats containing P. ilicifolia seeds were collected in burned and unburned chaparral. We counted seeds in each scat and noted whether endocarps had tooth punctures or rattled audibly when shaken. For comparative germination trials, we also collected fruits and seeds directly from mature shrubs. In the laboratory, following a cold-moist stratification period, seeds were assessed for germinability. In the field, we compared desiccation rates and germinability of seeds from bear scats and freshly collected seeds. We compared rates of moisture loss and germination for seeds subjected to several different conditions, including 25 °C (room temperature), 30 °C and 65 °C (in the lab), and placement on exposed soil in a burn area (in the field), where midday temperatures were approximately 45 °C but likely much higher by late afternoon. Prunus ilicifolia seeds collected from bear scats were largely undamaged; the vast majority of these seeds germinated. In some cases, germination rate was higher for seeds from scats than for seeds from intact fruits. Several results indicate that desiccation is an important cause of reduced germinability. First, seeds that rattled audibly germinated poorly; and the louder the rattle, the lower the germination percentage. Second, seeds (both fresh and those from bear scats) placed in the field under protective screens had greatly reduced levels of germination (an 84% decline) after only 7 days. Third, seeds dried in the lab, even at relatively moderate temperatures, showed a decline in germination with seed moisture loss. The addition of high temperatures accelerated this decline in germination. We discuss the relevance of heat and desiccation sensitivity of seeds dispersed by bears to successful seed germination in burned and late-seral mesic and xeric chaparral.

Tree recruitment is rare and oak populations are declining in many Mediterranean-climate oak savannas. Factors affecting acorn production, seedling establishment and initial seedling growth have been much studied in short-term experiments. However, fecundity and early survival rates have not been placed in a demographic framework that also considers sapling survival and growth and adult tree mortality. We use matrix models and life table response experiments to analyze long-term experimental and observation data on California valley oak (Quercus lobata Nee) near its southern range limit in Santa Barbara County to answer three questions: (1) How sensitive is oak population growth rate to variation in acorn production and initial seedling establishment vs. sapling survivorship and growth? (2) How do mammalian consumers-specifically, cattle, deer and rodents-impact valley oak seedling establishment and sapling survival and growth? (3) Can vertebrate consumers account for the observed population decline of valley oak in savannas in its southern range? We find that population growth is far more sensitive to consumer-mediated variation in sapling survivorship and growth than to variation in fecundity or seedling establishment and that consumers exert strong influence on the demography of the species. Deterministic, finite population growth rate (lambda) is <1 for unprotected plants and for plants that are protected from cattle but still exposed to mule deer and rodents. Population growth rate increases to 1.03 with protection from both cattle and deer, mainly because plants are able to quickly reach the tree layer when they are protected from ungulate browsing. Population growth rate jumps to 1.15 for plants protected from both ungulates and rodents as a result of increased survivorship and height growth of established individuals and because of increased seedling establishment during the first year. Our experimental findings are consistent with observed patterns in natural populations in Santa Barbara County, where tree recruitment is rare in both grazed and non-grazed savannas but more common in areas such as roadsides that are refuges from cattle and deer.

We review published studies on the demogrophy and recruitment of California oak trees and focus on the widespread dominant species of the foothill woodlands, Quercus douglasii, Q. lobata, and Q. agrifolia, to ascertain the nature and strength of evidence for a decline in populations of these species. The vast majority of studies have been of short duration (less than three years), focused on the acorn and seedling life stages, and conducted at few locations within each species' geographic range. We summarize the extensive body of research that has been conducted on the biological and physical factors that limit natural seedling recruitment of oaks. The oak "regeneration problem" has largely been inferred from current stand structure rather than by demographic analyses, which in part reflects the short-term nature of most oak research. Mien viewed over longer periods of time using field surveys or historical photos, the evidence for a regeneration problem in foothill oaks is mixed. Q douglasii shows very limited seedling or sapling recruitment at present, but longer term studies do not suggest a decline in tree density, presumably because rare recruitment is sufficient to offset low rates of mortality of overstory individuals. Q. agrifolia appears to be stable or increasing in some areas, but decreasing in areas recently impacted by the disease Phytophthora ramorum. Evidence from the few available studies is more consistent in suggesting long-term declines in foothill populations of Q lobata. Longterm monitoring, age structure analysis, and population models, are needed to resolve the current uncertainty over the sustainability of oak woodlands in California.

We quantitatively compared phenology and water relations of a fully deciduous shrub, Styrax officinalis, and an evergreen shrub, Arctostaphylos glauca, in shared microsites in a sandstone outcrop in southern California during a multi-year drought. Pre-dawn xylem pressure potentials, Ψpd, were similar for the two species during most months of 2 years, but occasional differences and watering experiment results suggest S. officinalis may have phreatophytic roots that tap water in deep rock cavities and joint traces, while A. glauca may have primarily shallow roots. Neither species varied in maximum or minimum Ψpd between years of very different rainfall totals. Twig elongation and leaf production of S. officinalis began earlier during spring, and its leaves matured more quickly and more synchronously than A. glauca. Leaves lived a mean of 180 days for S. officinalis and 849 days for A. glauca. Leaf life spans varied among years in both species. S. officinalis leaf senescence occurred mostly in August and September and was not discernibly related to Ψpd or drought avoidance. A. glauca leaf senescence occurred throughout the year, but especially coincided with leaf production. In A. glauca most senescence occurred at the beginning of a leaf cohort's third growing season, but numbers of retained older leaves increased during 4 years of drought. Timing of twig elongation and leaf and flower production appeared to be related to current rainfall, but amount of twig growth and numbers of leaves and flowers produced appeared to be related to rainfall of the previous year or years for S. officinalis and, more complexly, A. glauca. Because of an interrupted pattern of flower production, number of flowers produced by A. glauca may be responsive to rainfall amounts during two environmental periods. Morphological differences, including much higher above ground allocation, many more leaves/twig, lower allocation to stem mass, and longer duration of leaves/year in A. glauca, are probably responsible for A. glauca having >6 times more above ground biomass per plant than S. officinalis. During the multi-year drought S. officinalis changed little, indicating either superb adaptation or growth pattern rigidity, while A. glauca underwent extensive phenological and morphological changes, indicating either stress or adaptive flexibility.

We investigated California black oak (Quercus kelloggii Newberry) acorn dispersal by rodents and birds in the months after a stand-replacing fire in a mixed conifer forest in the San Bernardino Mountains of southern California, USA. The objective of this study was to compare scatter-hoarding in a high-severity burn to that in an unburned forest. In the fall of 2007, we placed 600 magnet-bearing acorns under trees in the unburned area. Of the 600, we recovered 77 (13%). Dispersers moved acorns an average distance of 5 m and buried them to an average depth of 30 mm. By spring of 2008, 90% of the cached acorns were missing. In the high-severity burn, we recovered 59 (9.8%) of the 600 acorns placed under top-killed oaks; these had been scatter-hoarded an average of 5.27 m from the source plots and buried an average of 22 mm. By spring of 2008, 55% these acorns were missing, and many of those that we relocated had been re-cached in new locations. Our results suggest that scatter-hoarding of acorns may be a common phenomenon after fire, and likely plays an important role in seedling recruitment.