From the collection of the j f o Prelinger h v Ijibrary 6 t San Francisco, California 2006 Released from Cr^nbrook htsHtute of oden SPIDERS THE NEW ILLUSTRATED NATURALIST EDITORIAL ADVISORY BOARD WILLIAM BEEBE, Sc.D., LL.D. Director, Department of Tropical Research New York Zoological Society AUSTIN H. CLARK Smithsonian Institution ROBERT CUSHMAN MURPHY, A.M., ScD. Chairman, Department of Birds, American Museum of Natural History FAIRFIELD OSBORN President, New York Zoological Society President, Conservation Foundation THE NEW ILLUSTRATED NATURALIST SPIDERS BY WILLIS J. GERTSCH, PH.D. Associate Curator, Department of Insects and Spiders, American Museum of Natural History D. VAN NOSTRAND COMPANY, INC. TORONTO NEW YORK LONDON NEW YORK D. Van Nostrand Company, Inc., 250 Fourth Avenue, New York 3 TORONTO D. Van Nostrand Company (Canada), Ltd., 228 Bloor Street, Toronto 8 LONDON Macmillan & Company, Ltd., St. Martin's Street, London, W.C. 2 COPYRIGHT, 1949 BY D. VAN NOSTRAND COMPANY, INC. Published simultaneously in Canada by D. VAN NOSTRAND COMPANY (Canada) LTD. All Rights Reserved This book, or any parts thereof, may not be reproduced in any form without written per- mission from the author and the publishers. Produced in collaboration with Chanticleer Press, Inc. PRINTED IN THE UNITED STATES OF AMERICA Preface OPIDERS MAKE UP A CONSIDERABLE portion of the animal life of the vast and diversified land that is North America. That general knowledge of them is relatively meager must be attributed to the circumstance of size, rather than to inferiority in either importance or genuine interest. By means of size, and also of sound, birds, mammals, and vertebrate animals mo- nopolize the stage and divert attention. Yet only a slight change in perspective will bring into view a microcosm of tiny creatures that, hidden away in leafy jungles or unseen in miniature forests under foot, live lives of unbelievable strangeness and complexity. To bring this microcosm into sharp focus for the general reader is the prime purpose of this book. Our American spider heritage is a large and diversified fauna commensurate in importance with the age and size of the continent itself. Proclaiming this heritage is a large and rewarding body of literature created by students during more than one hundred and fifty years of enthusiastic devotion. At the beginning one would mention the name of John Abbot, who, as early as 1776, began the study of spiders and other animals in the region around Savannah, Georgia. It is to be regretted that his fine paintings and accompany- ing notes were never published, as were those of the birds, butter- flies, and moths for which he became justly famous. Thereafter, with Nicholas Marcellus Hentz, whose first contribution appeared in 1821, began a line of investigators (H. C. McCook, T. H. Mont- gomery, G. W. and E. G. Peckham, J. H. Comstock, and J. H. Emerton, to mention only a few) which has terminated in that out- standing living American devotee of Arachne, Alexander Petrunke- vitch, and in a growing circle of younger workers. The contribution of Americans to world araneology has been a striking one, but we have profited in even greater measure by the energy and genius of students from other lands, foreigners in language only. vi PREFACE Our debt to the past is a very great one, and credit for our (often presumed) deeper insight into the Araneae must to a considerable extent go to the accumulation of information marshaled by the pioneers. The facts brought together in this book are borrowed largely from a fund of information available to all arachnologists, and, while they reflect commendable knowledge, at the same time they reveal comparative ignorance of much in the lives of the spin- ning creatures. It is therefore the author's hope that this book will, in addition to its other purposes, act as a stimulus to those eager to unearth the many details still unknown. Most spiders are difficult subjects that try the patience and tech- niques of photographers. It is thus particularly gratifying that an excellent and representative collection of photographs was available for use in this book. On the many colored and black-and-white plates are depicted graphically the forms, patterns, and handiwork of some of our commonest and most interesting spiders, almost all from living subjects. To those who have offered their photographs, many of them associates and personal friends, I extend my sincere thanks and further express my admiration for their splendid work. One of the contributors, George Elwood Jenks of Los Angeles, died before the completion of this book, leaving behind distin- guished pictorial explorations of the lives of spiders and their enemies as a monument to his enthusiasm. To my friend and col- league, Walker Van Riper of the Denver Museum of Natural His- tory, I offer my special gratitude. In addition to placing his valuable albums in my hands for use without reservation, he has aided ma- terially in securing photographs of the subjects most needed. Fi- nally, it is a privilege to acknowledge the contribution made by Dr. B. J. Kaston of Connecticut State Teachers College at New Britain, who, in spite of preoccupation with other work, found time to read and criticize a large portion of this book. All the sugges- tions he has made, which reflect his broad training in biology, have resulted in material improvement of the manuscript. W. J. GERTSCH Contents CHAPTER PAGE PREFACE v LIST OF ILLUSTRATIONS IX 1. Introducing Spiders i 2. The Place of Spiders in Nature 1 1 The Life of the Spider 28 4. Silk Spinning and Handiwork 52 5. Courtship and Mating 68 6. The Evolution of Spiders 99 7. The Tarantulas 107 8. The Cribellate Spiders 137 9. The Aerial Web Spinners 157 7 The Hunting Spiders 193 11. Economic and Medical Importance 236 12. The North American Spider Fauna 255 GLOSSARY 267 BIBLIOGRAPHY 271 INDEX 273 Vll List of Illustrations FOUR-COLOR ILLUSTRATIONS PLATE FACING PAGE 1. Orb web covered with dew 2 2. Orange Argiope, Argiope aurantia, in web 3 3. Crab spider, Misumena calycina, dropping on dragline 16 4. Spider Relatives a— Solpugids of the family Eremobatidae 17 b— Scorpion, Hadmrus hirsutus, stinging tarantula, Aphonopelma . 17 5. A humped orb weaver, Aranea gemmoides, on egg sac 32 6. Black Widows a— Black widow, Latrodectus mactans, in web 33 b— Black widow, Latrodectus mactans, ventral view 33 7. Egg Sacs a— Opened egg sac of orange Argiope, Argiope aurantia 48 b— Egg sac of shamrock orb weaver, Aranea trifolium 48 8. Cluster of baby orb weavers, Aranea, preparing to disperse 49 9. Crab spider, Misumenoides aleatonus, on flower 64 10. Southern Spiders a— Huntsman spider, Heteropoda venatoria 65 b— Silk spider, Nephila clavipes 65 11. Black Widow, Latrodectus mactans, with prey 80 12. Tarantulas a— Tarantula, Aphonopelma, and tarantula hawk 81 b— Tarantula, Aphonopelma, and tarantula hawk 81 13. Tarantulas a— Portrait of a tarantula, Aphonopelma 94 b— Side view of a tarantula, Aphonopelma 94 14. Orb Weavers a— Banded Argiope, Argiope trifasciata, in web 95 b— Spiny-bodied spider, Gasteracantha cancriformis, on leaf 95 15. Purse web of Atypus abboti against tree 1 10 1 6. Burrow of Folding-Door Tarantula, Antrodiaetus 1 1 1 a— Door open 1 1 1 ix x LIST OF ILLUSTRATIONS PLATE FACING PAGE b— Door half open 1 1 1 c— Door closed 1 1 1 17. Black widow, Latrodectus mactans, with egg sac 124 1 8. Shamrock orb weaver, Aranea trifolium, on flower 125 19. Banded Argiope, Argiope trifasciata, with swathed prey, dorsal view FIG. 4 20. Banded Argiope, Argiope trifasciata, ventral view 139 21. Orange Argiope, Argiope aurantia, in web, side view 154 22. Spiny-bodied spider, Micrathena gracilis, spinning 155 23. Orb Weavers a— Shamrock orb weaver, Aranea trtfolium 170 b— The garden spider, Aranea diadema 170 c— Orb weaver, Neoscona 170 d— Orb weaver, Neoscona, on leaf 170 24. Wolf spider, Geolycosa missouriensis, at mouth of burrow FIG. 5 25. Wolf Spiders a— Wolf spider, Geolycosa turricola, side view 182 b— Burrow of wolf spider, Geolycosa, in grass 182 26. Grass spider, Agelenopsis, on egg sac 183 27. Crab spider, Misumena calycina, on flower 198 28. Crab spider, Xysticus gulosus, with prey 199 29. Jumping spider, Phidippus mineatus, side view 214 30. Jumping spider, Phidippus cardinalis, on flower 2 1 5 31. Green lynx spider, Peucetia viridans, and nest 230 32. Jumping spider, Phidippus, dorsal view 231 GRAVURE ILLUSTRATIONS I. Banded Argiope, Argiope trifasciata, swathing a grasshopper 20 II. Orange Argiope, Argiope aurantia, with swathed prey 2 1 III. Female bolas spider, Mastophora cornigera, with re- cently emerged brood, including some adult males 28 A symmetrical orb web of a mountain orb weaver, Aranea aculeata 28 Meshed web of Dictyna on dried weed 28 LIST OF ILLUSTRATIONS xi PLATE FACING PAGE IV. A Jumping Spider, Phidippus audax, and its Dragline a— Preparing to leap 29 b— Leaping 29 V. A Juvenile Jumping Spider, Phidippus, On A Thin Toothpick, Prepares To Fly a— Orienting in response to breeze, secured by dragline 52 b— Ballooning threads stream from spinnerets 52 VI. Courtship and Mating In The Black Widows, La- trodectus mactans 53 a— The cautious approach of the small male 53 b— The mating 53 VII. Black Widows, Latrodectus mactans a— The male after mating is occasionally, as here, killed and eaten by the female 60 b— A female in her tangled snare with long-legged spiders, P silo chorus 60 VIII. Relatives of Spiders a— A desert solpugid (Eremobates) 61 b— A giant-tailed whip scorpion, Matigoproctus giganteus 61 IX. Spider Relatives: Harvestmen on aphis-covered rose shoots 84 X. Trap-Door Spider, Bothriocyrtum calif ornicum a— Molting. Carapace and chelicerae freed FIG. 2 b— Molting. The shed skin FIG. 2 c— Cradle of eggs in burrow FIG. 2 XI. California Trap-Door Spider, Bothriocyrtum call- jornicum a— Exposed burrow 90 b— Male 90 c— Cork-door nest held open 90 XII. California Trap-Door Spider, Bothriocyrtum cali- fornicum a— Capturing a ground beetle 91 b— Lifting the cork lid 91 XIII. Female purse web spider, Atypus bicolor 114 XIV. A Western Trap-Door Spider, Aptostichus, Dorsal View of Male FIG. 3 A Mexican Trap-Door Spider, Eucteniza xii LIST OF ILLUSTRATIONS PLATE FACING PAGE a— Surprised in its burrow FIG. 3 b— Exposed burrow FIG. 3 XV. Male Tarantula, Aphonopelma a— Clambering over stone 120 b— Portrait 1 20 XVI. Tarantula, Aphonopelma a— Female on desert soil 1 2 1 b— Web-covered entrance to burrow 121 c— Female and egg sac in exposed burrow 1 2 1 XVII. Tarantula, Aphonopelma, and Tarantula Hawk a— The tarantula assumes a defensive attitude 142 b— The wasp inserts its sting 142 c— Pulling the bulky prey to prepared burrow 142 XVIII. Silver Argiope, Argiope argentata a— Female and pygmy male 143 b-Egg sac 143 Egg Sac of Orb Weavers a— Banded Argiope, Argiope trifasciata 143 b— Humped orb weaver, Aranea gemmoides 143 XIX. Long-Legged Cellar Spiders, Pholcus phalangioides a— Male and female, with eggs, in tangled web 150 b— Female holding mass of recently hatched young 1 50 XX. A Comb-Footed Spider, The Black Widow, La- trodectus mactans. Captures A Jerusalem Cricket a— The spider approaches as the cricket touches the capture threads 151 b— Nooses of swathing film are combed over the leg 151 XXI. A Comb-Footed Spider, The Black Widow, La- trodectus mactans, Captures A Jerusalem Cricket c— Tiny fangs inject the venom 172 d— The bulky insect is lifted above the floor 172 XXII. A female humped orb weaver, Aranea gemmoides% clinging to a plant 173 A female humped orb weaver, Aranea gemmoides, hanging in the hub of her orb web 173 A fisher spider, Pisaurina mira, with egg sac 173 LIST OF ILLUSTRATIONS xiii PLATE FACING PAGE XXIII. Mud Dauber, Mud Nest and Spider Prey 180 The Bolas Spider, Mastophora cornigera a— Portrait 180 b— The pendent egg sac, opened to show young 1 80 XXIV. Feather-foot spider, Uloborus americanus, with egg sac FIG. 6 A symmetrical orb web of banded Argiope, Ar- giope trifasciata FIG. 6 Female of tuberculate Cyclosa, Cyclosa turbinata, on egg string FIG. 6 XXV. Wolf Spiders a— A female Lycosa covered with young 202 b— Portrait of male, Pardosa milvina 202 c— Turret of burrow of Lycosa carolinensis 202 XXVI. Wolf Spiders, Lycosa a— With captured fly 203 b— With attached egg sac 203 XXVII. The Green Lynx Spider, Peucetia viridans a— Female and egg sac 210 b— Male . 210 XXVIII. A Fisher Spider, Dolomedes scriptus 2 1 1 Grass Spider, Agelenopsis. An immature male sits in its tunnel 2 1 1 Web of a grass spider, Agelenopsis, blankets the soil 2 1 1 XXIX. Hunting Spiders a— A giant crab spider, Olios fasciculatus 246 b— A crab spider, Misumenoides aleatorius 246 XXX. Hunting Spiders a— Male and female running spiders, Trachelas, in silken cell 247 b— Running spider, Chirac anthium incluswn, with egg sac 247 XXXI. Portrait of wandering spider, Cupiennius 262 A wandering spider, Ctenus, with egg sac 262 Portrait of jumping spider, Phidippus 262 XXXII. Jumping Spiders a— Phidippus formosus stalks a fly 263 b— Phidippus audax with bee fly 263 CHAPTER I Introducing Spiders SPIDER PREVIEW HIS BOOK TREATS OF THE SPIDERS of the United States and Canada and is concerned almost wholly with their habits and life histories, their morphology and peculiarities, and also with their numbers and kinds. Most of us know something about spiders, but few of us are aware of the vast numbers that exist and of the great diversity in appearance and habits of the spin- ning creatures. Yet even a limited acquaintance soon makes it evi- dent that spiders in many ways far outshine insects and lesser animals of much greater reputation. Thus it seems desirable that, at the very outset, a few of the striking peculiarities of the maligned spiders be enumerated. Insects have developed wings and on them have attained the most exalted place among the arthropods. Although a wingless creature of the earth and its plant cover, the spiderling can float its threads on the breezes and fly through the air, often reaching tremendous heights and sailing for long distances. This "ballooning" of spiders has been instrumental in distributing them into new colonizing areas at a rate not possible even for insects with their wings. The rigging of ships two hundred miles from the nearest land has been showered with tiny aeronauts riding on silken streamers. The spider can spin a line one-millionth of an inch in thickness, but most of its single lines are ten or twenty times as thick. This strand of silk is a line of great elasticity that will stretch one-fifth its length before break- ing, and of a tensile strength second only to fused quartz fibers. It is a line of such fineness that it is impossible to duplicate; it serves admirably as a marker in various surveying and laboratory instru- ments. An inveterate spinner during all of its life, the spider uses silk for so many different purposes that this material is the most 2 AMERICAN SPIDERS important thing in its life, the factor that has largely determined its physical form and dominant place in nature. Almost alone among the lesser creatures the spider prepares a trap to capture its prey. By their structure these traps are identified as tube webs, purse webs, sheet webs, tangled webs, and orb webs. Sometimes they are complex structures of very curious form. The orb web (PI. I and PL III) has long been a symbol of the spider in the mind of man, who sees in its shimmering lightness and intricate, symmetrical design a thing of wonder and beauty. Such esteem is well merited, for the orb web is the most highly evolved of all the space webs developed by the sedentary spiders. It repre- sents a triumph in engineering worthy of great mechanical inge- nuity and learning; yet it was arrived at by lowly spiders, which even by their most ardent supporters are credited with hardly a gleam of what is called intelligence. The ingredients of almost un- limited time, of moderate compulsion to irresistible change, and the stimulus of real advantages gained have contrived to produce the two-dimensional orb web from the seemingly wasteful tangle of threads that is its origin. Instinctively and blindly the spider has followed the long path leading to its symmetrical masterpiece. The orb weavers are virtually slaves of their webs and have wagered their future on the tenuous lines. Within the limits of their circum- scribed world they are supreme autocrats, but when brushed from their snares, many are clumsy, vulnerable creatures. In accomplishing the purpose of entangling flying insects, the web has served the needs of the spider admirably and at remarkably small cost. Only about an hour is consumed in spinning the average orb web, which, because of great damage to the lines, frequently is replaced every suitable night by the methodical spider. Yet within the orb- weaving group there are some members that have broken so completely with the past that they do not spin orb webs at all but have substituted an entirely different method of securing their prey. Instead of relying on the static but dependable round web, they spin a line, weight the end with a sticky drop of liquid silk, and hurl it much as the gaucho throws his bolas or the angler casts his line. One need not travel to the exotic tropics to find these bolas spiders; they live over most of the United States and even within large cities, seeming to prefer the trees of our formal parks. Close rela- tives of the bolas spiders live in Australia and Africa; one of these African cousins varies the casting procedure by spinning its line around like a whirligig. PLATE I Richard L. Cassell Orb web covered with dew PLATE 2 Orange Argiope, Argiope aurantia, in web INTRODUCING SPIDERS 3 The female bolas spider (Pis. Ill and XXIII) is a plump creature, about one-half inch long and equally wide, which sits placidly on a twig, simulating with considerable faithfulness a bud, a nut, a snail, or even a bit of bird dung. What about her mate? He is an insignificant atom no larger than the head of an ordinary pin. Pre- cociously developed, he walks out of the egg sac fully mature, along with sisters his own size who are just beginning their life and must wait weeks and increase tremendously in size before they become sexually mature. Spiders and their relatives are ancient animals; they were among the first creatures to leave the waters for a life on land. Some mod- ern spiders seem to be only thinly masked replicas of creatures that were living in the northern hemisphere during the remote Paleozoic Era, when the coal measures were still in infancy. Although more generalized than the commoner true spiders, the tarantulas and their kin have become specialists in their own fashion, and have devised new and extraordinary ways of living in a world of competition. The purse-web spiders live in a long silken tube closed at both ends, and have developed long fangs with which they impale insects that walk over their cylinder by biting through it. The burrowing taran- tulas of the genus Antrodiaetus ensure privacy in their burrow home by pulling two flaps of silk, which fit like folding doors, over the entrance. The trap-door spiders are accomplished burrowers and cap the opening to their chamber with a hinged trap-door. One of the strangest trap-door spiders is Cyclocosmia, which has an ab- domen hardened and rounded behind to form a plug with which it at one time was reputed to close its burrow. Among the vagrant tarantulas are some that have become verita- ble giants far exceeding most insects in bulk and rivaling in size even the great black scorpions of Africa. Armed with long, strong fangs, they are able to kill with ease and feed on frogs, toads, and lizards, and also to subdue and eat rattlesnakes and other larger animals. Some of the arboreal tarantulas are known to kill small birds, and have gained one of their common names of "bird spiders" from this activity. Longer-lived than any terrestrial invertebrate are some of the great hairy tarantulas, which do not become sexually adult until eight or nine years old and are known to live thirty years. Among the true spiders are the diurnal jumping spiders; these actively pursue their prey over the ground and on plants. Special tufts of adhesive hairs on the tarsi allow them great freedom of movement on precipitous surfaces, and, aided by the keenest eye- 4 AMERICAN SPIDERS sight of all spiders, they emulate the carnivores in stalking their prey. Their stout bodies and legs are gaily colored and bedecked with tufts of bright hairs, pendant scales, and curious spines. Gleam- ing with their iridescent scales like jewels in the sun, they rival the gaudiest insects. During courtship dances, the little males caper and posture before the females in such manner as to display their bril- liant ornaments to best advantage. In the petals of many kinds of flowers hide stubby little crab spiders which, simulating the assasin bugs, seize flying insects that visit the blossoms for nectar. In keeping with their role of decep- tion, they change from white to yellow, or vice versa, to conform with their background. All spiders breathe air through orifices on the ventral side of the abdomen. In spite of their air requirements, many have adopted an amphibious life and stay under water for periods of variable length. Some live in little waterproof chambers spun in holes in coral rock that are covered over during high tide. Most extraordi- nary of all is the water spider, Argyroneta of Eurasia, which is able to swim about and live for weeks in the fresh water of streams and ponds in a domicile that resembles a small diving bell. This spider carries air bubbles beneath the surface to its retreat, which is an- chored to aquatic plants by silk lines, and keeps a supply of air imprisoned in the silken chamber. Its prey consists of small aquatic animals, which it captures in the stream. Even the eggs are laid and the family hatched out under water in the security of the nest. Among American amphibious spiders are some of the fisher and wolf spiders, which run over the surface freely and dive into its depths where they stay for long periods. Occasionally small fish or amphibians are caught by the large fisher spiders of the genus Dolo- medes. The sexual characteristics of spiders are especially interesting. In both sexes the genital opening is a simple pore beneath the base of the abdomen through which emerge the spermatozoa or eggs. One would expect that during mating the male products would be transferred directly to the female by contact between these orifices or by means of an eversible intromittent organ. Instead, the male spider has transformed the claws on the ends of the pedipalpi (the leglike appendages lying on each side of the head in both sexes), into a complicated intromittent organ, comparable to a syringe or a hypodermic needle, and has modified and greatly enlarged the distal segments of the pedipalp to protect the organ and facilitate INTRODUCING SPIDERS 5 the pairing. These organs of the male, called palpi, have no internal connection with the gonads of the abdomen, so the semen must be transferred from the genital orifice to the palpi. To accomplish this the male spins a little sperm web, deposits a small globule of semen upon it, and then sucks it into the syringe in each of the palpi. The female has developed in front of the genital pore paired pouches for the storage of the semen, each unit of which is shaped to receive the corresponding palpus of the male. Since spiders are solitary, predaceous creatures, the male runs considerable risk in approaching his usually much larger mate, who may be only hungry and not ready for mating. Some males are killed because of early failure to diagnose the attitude of the female, or, after being successful in their suit, of not leaving the premises before the normal predatory instincts of the female again dominate her. Various routines have been devised by different groups of spiders to gain the recognition of the female and make possible a transfer of the semen in relative safety. In the bodies of spiders are found clues that give considerable insight into the racial history of the group. From lumbering ground creatures have come fleet runners on soil and vegetation, and trapeze artists that hang in midair on silken lines. In the variety and strange- ness of their forms, spiders surpass all comparable invertebrate groups. In color pattern, ornamentation, and brilliance they are on a par with any of the insects. Indeed, the vaunted brilliance of the morpho butterflies and of the birds of paradise is excelled by the iridescent variety of the jumping spiders of the tropics. Only the small size of spiders conceals their beauty and keeps them largely unknown. Finally, it should be noted that spiders have attained their pres- ent position without benefit of so-called intelligence. Endowed with incredibly complicated instincts, the spinning creatures per- form their marvels largely as automatons, and show only moderate ability to break the bonds of their behavior patterns. The baby orb weaver spins a perfect orb web soon after it leaves the egg sac, and thereafter scarcely changes it, except in size, during its whole lifetime. The mother spider encloses her eggs in a sac which, often beautifully designed, advertises the species to which she belongs, and then defends her precious burden against any assailant. Instinct plays a large role in every action of the spider and is the guiding principle throughout its life. AMERICAN SPIDERS GENERAL ATTITUDE TOWARD SPIDERS Spiders are seen in different lights by different peoples. Primitive men regard some spiders as bad, others as good, and most as having little importance or significance in their lives. To those that become important because of venomous or presumed dangerous character, they give special names. The chintatlahua of the Oaxaca Indians, the po-ko-moo of the Mewan tribe of California, and the katipo of the Maoris, all refer to similar spiders of the genus Latrodectus, which have long been notorious over much of the temperate and tropical world. Each people has a distinctive name for the brightly marked spiders known as "black widows." In addition, species re- sembling the virulent ones are regarded with suspicion and often endowed with the same venomous powers. This is a practical ap- proach, learned by trial and error, and tested in time by peoples who have close contact with the lesser creatures about them. There- fore it is not surprising that the beliefs of primitive peoples often have a firm foundation in fact. In the second category are some spiders that are good because their presence at certain hours, on specific occasions, in particular places, constitutes a good omen. A few are eaten with keen relish. Others are seen as wonderful creatures that produce marvelous webs overnight and have magical powers. To the American Indians the spider is a creature of mystery and power, which, though capable of trickery, duplicity, and even great evil, plays a benevolent and often potent role in many of their le- gends. The prowess of spiders in this folklore is based largely on their great skill as spinners, and to a lesser extent on the deadliness of their bite. To the Dakotah the orb web is a symbol of the heavens; the corners of the foundation lines point in the four direc- tions from which come the thunders, while from the spirals of the orb emanate the mystery and power of the Great Spirit. In Indian legend spiders are venerated for spinning silken lines of great strength on which some unfortunate is able to escape from des- truction. A youth, betrayed into sleep by the seduction of a woman, awakes on a precipitous cliff but lowers himself to safety on a line furnished by a spider friend. This same silken cord may also be a rope to the sky on which the dead mount to the new hunting ground, or the brave climb to wreak vengeance on the sky people. But more often it is a line from the sky to the earth on which the INTRODUCING SPIDERS 7 pursued can descend; it is on such a "sky rope" that the Algonkin maiden, fallen from grace as wife of the Morning Star, is sent back to earth. In many interesting myths of the Pueblos the main role in the Creation is assigned to the spider. According to the Sia Indians, in the beginning there was only one personage, a spider, living in a world sterile of life and lacking many material things. From each of two little packages possessed by the spider was conjured, in re- sponse to its magical singing, a woman. From the first woman thus created have descended all the Indians, and from the second all the other races of men. Some of the virtues attributed to spiders are industry, patience, and persistence. Well known is the legend of Robert Bruce who gained new courage by watching a spider finally reach its cobweb home after many unsuccessful attempts. In a delightful Cherokee myth the little spider appears as a successful agent when all other animals fail. In the beginning the world was cold. Then fire ap- peared on the earth, having been placed in a hollow tree on an island by thunder and lightning. The shivering animals gazed across the waters and resolved to secure the warmth of the fire for their own purposes. After consultation, the raven was dispatched to secure the bright embers, but was unsuccessful and soon returned with blackened feathers, which it wears to this day. One by one the birds, snakes, and other animals risked a trial, but all brought back only scars from the fiery furnace in the tree. Finally, the spider alone was left to brave the waters. She prepared herself by spinning a little tusti-bovsl of her silk, which was then fastened to her back. Skating across the surface of the water, she crept through the grasses to the site of the fire, caught a little ember in the tusti- bowl, and delivered the priceless jewel to the waiting animals. This successful venture is usually attributed to one of the amphibious wolf spiders, which drags its egg sac behind it, attached to the spinnerets. A legend of great antiquity is that of the Spider Woman of the American Southwest, who is credited with being the inventor of weaving and the teacher of all textile art to the various Indian tribes. She is an earth goddess and usually lives in a burrow deep in the soil with the Spider Man, her husband. According to Navajo legend, the art of blanket- and basket-weaving was brought to them by an unhappy Pueblo girl from Blue House, near Pueblo Bonito, who came to the hogans of the Navajos to earn her living. One day the 8 AMERICAN SPIDERS girl wandered far from the hogan and, attracted by a thin wisp of smoke, discovered a small hole in the earth at the bottom of which was an old woman spinning a web. It was the Spider Woman, who quickly invited the girl to enter her house and blew up the hole until it was large enough to accommodate her guest. Befriended by the kindly Spider Woman, the girl stayed several days and learned to weave the blankets and baskets that now distinguish the Navajo. The Pueblo girl then transmitted this weaving art to her adopted people, and along with it an admonition from the Spider Woman that to forestall bad luck a hole must be left in the middle of each article. In compliance with this request, the Navajo women left a spider hole in the middle of each blanket, like the entrance to the burrow of the Spider Woman. Even to this day the spider hole may still be found in the blankets and baskets of the Navajo. Its position and form are greatly changed and masked in deference to the wishes of persons who pay a better price for flawless examples. Needless to say, it is always present in the blankets of the old women, who do not care to risk the anger of the mythical Spider Woman and the threat she made to spin silken threads in their heads. In a number of legends, spiders are placed in an unfavorable light and are pictured as villains and murderers. Thus the Win- nebagos tell of the eight blind men who snared and killed people with long cords strung among the trees. Wash-Ching-Geka, the Little Hare, went among the evil creatures, incited them to quarrel- ing, and then poisoned the meat they were cooking. They ate of the meat and were soon dead, whereupon Wash-Ching-Geka dis- covered that they were in reality spiders. The duplicity of the spider is dwelt upon in the rhyme of the Spider and the Fly, and that theme also occurs in the Indian legends. Here the spider is often a rascal and excels as a trickster. The Zuni tell a very pleasing story of how "old tarantula" dupes a handsomely dressed youth and finally absconds with his prizes. The youth is persuaded to allow "old tarantula" to don his fine clothes so that he can appreciate how handsome he appears in the eyes of others. "Look at me now. How do I look?" asks the spider as he dis- plays the garments. The youth, finding the ugliness of the wearer somewhat detrimental to the appearance of the clothes, is not greatly impressed. The spider moves off a bit, and as dis- tance lends enchantment, or at least makes repulsiveness less INTRODUCING SPIDERS 9 obtrusive, the youth notes an improvement. Still a little farther off moves the spider, pretending that his only object is to gain the youth's approbation, but really intent on getting nearer and nearer to his burrow. At last he arrives at the entrance. "How do I look now?" asks the wily creature. "Perfectly handsome," replies the youth; but as he speaks the spider dives into the earth with the stolen finery.1 Many curious beliefs are current in various parts of the United States regarding spiders, and often they are contradictory. It is rather generally believed that killing a spider or a daddy-long-legs will bring rain, and that many cobwebs on the grass in the morning foretell clear weather. The color of a spider is frequently of much significance in these superstitions. Black ones are almost invariably bad, just as white ones almost certainly signify good luck, but oc- casionally the colors are reversed and assume the opposite attribute. Although in some cases they are thought to be unlucky, the appear- ance of spiders is usually supposed to signify good luck, bringing to the observer new clothes, gifts, money, or visitors. Spiders have gained notoriety by smaller effort than any other animals. The bad reputation of a few species has been magnified beyond reason and is now attached to all of them. There is a gen- eral belief throughout the United States, and probably over much of Europe, that the bite of any spider is poisonous. Public opinion has been influenced by tall stories from far places, by sensationalism in the newspapers, and by the natural prejudices of housewives who can be forgiven for wanting their rooms completely free of all crawling creatures. Spiders are for the most part small, and, because of their nocturnal habits, rarely intrude upon our notice. Much of the general aversion for them can be traced to teachings from par- ents and grandparents who early instill the young child with mis- information. The popular prejudice, which even finds expression in nursery rhymes, often amounts to a phobia. The squeamishness of grown men who "can't stand" spiders of whatever size contrasts most unfavorably with the nonchalance of small Indian boys who keep pet tarantulas on a string. A frank dislike of spiders because of their predaceous habits would put the whole business on a rational basis. The spectacle of insects being pounced upon, trussed up, crushed, and sucked dry is 1 H. F. Schwarz, "Spider Myths of the American Indian," Natural History, Journal of the American Museum of Natural History, Vol. 21 (1921), pp. 382-5. io AMERICAN SPIDERS one that prejudices us in favor of the underdog. But we have little dislike for other creatures, such as the ladybugs, which are quite as voracious. It is doubtful that people give sufficient heed to spiders to be affected by their rapacious methods; they are labeled nasty, crawly creatures in a completely irrational manner. CHAPTER II The Place of Spiders in Nature RELATIONSHIP TO OTHER ARTHROPODS T J.HE VAST ASSEMBLAGE OF ANIMALS comprising the phylum Arthropoda includes such familiar creatures as the crabs and lobsters, centipedes, millipedes, and insects, as well as the spiders and their multitudinous kin. Indeed, three fourths of the known animals of the world are arthropods and attest by their numbers, their variety, and their occupancy of every conceivable place in nature a degree of success not even closely approached by any other group of animals. Present in numbers conservatively esti- mated as beyond a million different species, they make up in vast populations what they concede to the vertebrates in size. Most of them are small, and because seven out of every ten kinds are insects, the average size is perhaps as small as a quarter of an inch. Indeed, it is perhaps to this small size, and to superior armament in the form of a tough but light external covering, that they owe their domi- nance in the world. The arthropods have their bodies encased in a stiffened outer covering, or exoskeleton, and completely lack the type of internal skeleton present in the vertebrates. The integument is made imper- meable to liquids and gases and kept hard and tough by the presence of amber-colored substances called sclerotin and chitin. Between the body segments and the joints of the appendages the cuticle is not so strongly impregnated with sclerotin and remains soft and pliable, allowing movement of the legs and other articulated seg- ments of the body. The problem of growth in size has been solved in the arthropods by their shedding the rigid outer skeleton at rather definite intervals, a process called molting. All the increase ii 12 AMERICAN SPIDERS in size of the carapace and appendages, and often of the abdomen as well, must take place immediately following molting when the integument is still soft. One characteristic of all the arthropods is the fact that their bodies are divided transversely into numerous well-marked rings or segments (in some cases most indications of segmentation are lost). The segments in front, which go to form the guiding center of the animal, are usually dissimilar and so greatly modified and fused that their exact limits are obscured. Thus, the head in one group is not necessarily the same as the head in another; it may be composed of more segments or carry more appendages, and the appendages of the same segment may be vastly different. From primitive append- ages have been derived mouth parts, swimmerets, legs, spinnerets, antennae, and many other organs. They are used for feeding, swim- ming, running, silk-spinning, mating, and for sensory perception. The hind portion of the animal, which is called the abdomen, is like- wise not the same in all the arthropods. In the centipedes and milli- pedes it is a multisegmented trunk, provided with numerous jointed legs, in some instances nearly two hundred pairs. In the insects the abdomen completely lacks appendages except at the caudal end. In spiders the only abdominal appendages are the spinnerets. With such marked difference in the external form of the Arthro- poda as compared with vertebrates, it is not surprising that the internal anatomy should also be quite distinct. The various systems for carrying on living, such as those for digestion, respiration, excre- tion, and reproduction, show marked differences. In the horseshoe crabs and most of the crustaceans, the respira- tory organs are external gills, which aerate the blood by absorbing through their delicate walls the oxygen and other gases dissolved in the water. Whereas most of the other arthropods long ago aban- doned an aquatic life, some individualists among the insects have secondarily returned to its security during parts of their life, but not before they devised new means of living. Respiration in the land arthropods is effected by means of internal aerating chambers called book lungs and tracheae, or, less frequently, by breathing directly through a soft outer covering. The book lungs of the arachnids are closely packed sheets of body surface bound together like the leaves of a book, to give the maximum surface for aeration. The tracheae in the arachnids are small tubes that lead into the body and sometimes ramify to form complex systems. In the myri- apods and insects, the air is conducted directly to the tissues by THE PLACE OF SPIDERS IN NATURE 13 means of tracheae which, however, are dissimilar to those of the arachnids and develop in a different way. Although simple diffu- sion through the skin or into the body by means of the tracheae often is sufficient in small, inactive arthropods, in large and more active forms some sort of breathing takes place, usually through rhythmical movements of parts of the body by special muscles of the abdomen. The type of circulatory system in the arthropods is a specialized one, seemingly highly efficient within the size limits of these crea- tures. Instead of closed tubes that carry the blood to every part of the body and ramify in great profusion to reach all the tissues, the system is at least in part an "open" one. The place of the veins is taken by expansive channels, or sinuses, filled with blood, in which the organs and tissues are bathed. In a large pericardial sinus lies the heart, which expands to allow the blood to enter the pumping vessel through paired lateral openings, and contracts to send the blood coursing through the arteries to all parts of the body. The blood is ordinarily a clear liquid in which are suspended numerous pale blood cells. A disadvantage of having the internal organs bathed in blood is the seriousness of accidental rupture of the outer covering. Any breaking of the body wall might prove fatal to the creature, since the blood would quickly drain from the body, but the tough exoskeleton guards against this. Injury to an appendage could also be fatal, but in many arthropods the injured member is removed by breaking it off (a process called autotomy) at a point where healing is rapidly accomplished. The digestive system is a tube that extends from one end of the body to the other, and is often subjected to various types of elabora- tion by coiling and compounding to increase the amount of absorp- tive surface. In the spiders and their relatives, this is accomplished by extending arms in many directions from the main tube. A con- siderable diversity exists among the arthropods as regards the de- tails of the digestive system, but all are alike in having a foregut and hindgut derived from the infolding ectoderm, and an expansive midgut, in which absorption is accomplished by means of the en- zyme-producing epithelium. The foregut in spiders is modified to pull in the liquid food. It consists of a pharynx into which the small mouth opens, an esopha- gus, and a so-called sucking stomach. The former are rather simple tubes, but the sucking stomach is an enlargement behind the esopha- gus supplied with powerful muscles on its four sides. When these i4 AMERICAN SPIDERS contract, they increase the size of the stomach and there results a strong sucking action that pulls the predigested food into the mid- gut. All the absorption occurs in the midgut, which is notable for a series of large blind sacs in the cephalothorax extending as four thick arms on each side, and voluminous glandular extensions from the main digestive tube in the abdomen. The hindgut provides a channel of egress for the fecal material, a thick, whitish liquid that is accumulated in a large bladderlike sac called the stercoral pocket, and voided through the anus. In addition to the tubular Malpighian vessels opening into the hindgut, which serve as excretory organs, spiders have a pair of coxal glands located opposite the coxae of the first and third legs, and these discharge their products through tiny openings behind the coxae. It is believed that the coxal glands are modified nephridia, the primitive excretory organs of earthworms and other animals, and that from similar glands in other parts of the animal have de- veloped the various silk glands and perhaps the poison glands as well. The activities of the arthropods are governed by a nervous system quite different from that found in the vertebrates. In the simpler forms it consists of a double nerve cord lying below the alimentary tract, which is enlarged in each segment to form a center or ganglion, from which lesser nerves arise. The most anterior pair of ganglia lie above the pharynx, and, joined to the pair immediately behind and below the pharynx by nerve connections, is called the brain. A very considerable modification of this generalized condi- tion is to be seen in most of the spiders, which have contained within the cephalothorax all the central nervous system. The ganglia in the cephalothorax have been consolidated into a single mass around the esophagus and below the digestive system. From the dorsal brain arise the nerves for the eyes and the chelicerae, and from the lower mass large nerves go to the appendages and back into the abdomen through the narrow pedicel. Sensation from the external environment is communicated to the central nervous system by means of structures called receptors. The most obvious ones are the eyes, which are remarkable organs in some insects but by comparison very feebly developed in spiders. Also very poorly represented in the arachnids are receptors for chemical stimuli, such as smell and taste, and perhaps the former sensation, as it is understood in vertebrates, is not even present in spiders. The receptors for touch are numerous and varied in the THE PLACE OF SPIDERS IN NATURE 15 Arachnida, and it is through their stimulation that these animals are best able to know their environment. THE NEAR RELATIVES OF SPIDERS The spiders and spiderlike animals belong to the class Arachnida, one of the major divisions of the Arthropoda. They differ at sight from most other arthropods in completely lacking visible antennae, the sensory appendages on the heads often appropriately called "feelers." Although frequently confused with insects because of similar size and general superficial appearance, the arachnids are not close relatives of these creatures, which have only three pairs of legs and have developed wings. All adult arachnids have four pairs of legs, except in rare instances, and they never have wings. Important and interesting in their own right are the arachnid relatives of spiders, such as the scorpions, harvestmen, and mites, which in this book can be mentioned only briefly in passing. Some were among the first animals to crawl out upon the land and adjust themselves to a terrestrial existence. And, rinding the land a most suitable zone for their development, almost none have returned to the water to live even part of their lives, as have many insects. A few of the mites have invaded both fresh and salt water, where they largely live parasitically on the bodies of aquatic animals. Each of the major groups, or orders, of the Arachnida has de- veloped a distinctive form, and the various types are quite familiar to most people. The members of the following four orders have the abdomen broadly joined to the cephalothorax by a thick waist: Order Scorpiones the Scorpions Order Pseudoscorpiones the Pseudoscorpions Order Opiliones the Harvestmen Order Acari the Mites The remaining orders of the Arachnida have the abdomen narrowed and attenuated in front to join the cephalothorax by a narrow waist: Order Solpugida the Solpugids Order Ricinulei the Ricinuleids Order Pedipalpi the Whip Scorpions Order Palpigradi the Micro- Whip Scorpions Order Araneae the Spiders 16 AMERICAN SPIDERS Every one of these living orders of arachnids occurs within the limits of the United States. A few interesting peculiarities of each should be noted. Scorpions. The scorpions are the most primitive members of the land arachnids, and also the oldest, being known from Silurian fossils that have an age of about four hundred million years. Among the oldest is a species from fossil beds at Waterville, New York, which was named Proscorpio osborni, and which was perhaps the first animal to adjust itself to a land life in North America. This ancient creature had no tarsal claws, and perhaps had not completely divested itself of the external gills that characterize the related, extinct eurypterids. The most obvious characteristic of the scorpion (Plate 4) is the invariable presence of a poisonous sting on the end of the ab- domen, which is narrowed to form an elongate tail. In life, the tail is curved over the back, and the spinelike sting is directed forward, always in position to attack its prey. The sting is generally used in conjunction with the great pedipalpi, which are developed as pin- cers to grasp and hold the victim. The venom of most scorpions is capable of causing mild to severe local reactions. A few species are known to cause pronounced neurotoxic reaction in man and warm-blooded animals. Two species of Centruroides occur in Ari- zona and are more notorious than the black widow for the virulent nature of their sting, which often is serious or fatal in children. Scorpions produce living young that mount the back of the mother and stay there until after their first molt, usually for a week or more. During this time they do not feed, but rely for sustenance upon the food stored in their bodies. The story that these little creatures, weakly armed with tiny chelicerae, feed upon the body juices of the mother, is a figment of some fertile imagination. An- other fable is the belief that scorpions commit suicide by stinging themselves when they are helplessly cornered or surrounded by a ring of fire. Pseudoscorpions. The pseudoscorpions are so named because of their superficial resemblance to true scorpions. They have the same enlarged pedipalpi terminating in pinching chelae, but the seg- mented abdomen is broadly rounded behind and is without trace of whip or tail. The largest species are scarcely more than one fourth inch in length, and most of the others are much smaller. PLATE 3 Walker Van Riper, Colorado Museum of Natural History Crab spider, Misumena calycina, dropping on dragline > > a. Solpugids of the family Eremobatidae %? «£ Richard L. Cassell rtr^fc -T,r ** *• * W*£fc,j| I A 4 **gj^ Richard L. Cassell b. Scorpion, Hadrurus hirsulus, stinging tarantula, Aphonopelma SPIDER RELATIVES THE PLACE OF SPIDERS IN NATURE 17 They live under stones, in moss, leaves, or debris on the ground, under the bark of trees, in the nests of bees, ants, and termites, and often in the dwellings of man. Many are found only in caves. One of the better-known species, the large, cosmopolitan Chelifer can- croides, lives in houses and shelters of man all over the world. The food of pseudoscorpions is believed to consist of mites, psocids, springtails, and other tiny insects, which are grasped with strong claws and perhaps anesthetized by venom from tiny glands in the chelicerae. Along with the spiders and some of the mites, the pseudoscorpions share the ability to produce a kind of silk. It comes from glands that are probably homologous with those that in spiders produce the venom to subdue prey. During periods when they are relatively helpless, such as when the female is distended with eggs, or when molting, they spin silk copiously and enclose themselves in wonderfully constructed nests or retreats. As is true of most arachnids, the pseudoscorpions have very poor vision, and frequently eyes are lacking altogether. The numerous sensory hairs on the pedipalpi and on other parts of the body take the place of eyes. The pseudoscorpions frequently attach themselves to the bodies of such insects as flies and bettles, and are thus transported quickly from one locality to another. Harvestmen. Familiar to most people because of the great length and thinness of their legs, the harvestmen, or daddy-long-legs, scarcely need introduction. Though often confused with spiders, to which they have a certain resemblance, they can always be dis- tinguished from their spinning relatives by the body, which has the cephalothorax and abdomen broadly joined to form a single unit. In this respect they are similar to the mites but differ from them in having the abdominal portion with well-marked segments. Most of the harvestmen (Plate IX) found in the temperate zone are active creatures that run rapidly on stilt-like legs, which they shed readily when in danger of being caught. They often congre- gate in considerable numbers on vegetation or on the trunks of trees, and are especially noticeable during the harvesting season, a fact that has inspired the common name. The harvestmen seem to feed largely on dead insects, but are also known to kill small ones for food, and to suck juices from various soft fruits and vegetables. The long-legged harvestmen are partially replaced in the warmer parts of the United States and in the tropics by shorter-legged spe~ i8 AMERICAN SPIDERS cies, which tend to be less active and frequently are quite sluggish. Many of these are bizarre animals that have beautifull sculptured bodies, often set with strangely shaped spines, and short legs, fre- quently armed with spines and processes. Many of them occur in caves in our southern states, where they have developed some un- usual types. Mites. Mites far surpass the other arachnids in numbers and economic importance. Most are minute reddish creatures with un- segmented, ovoid bodies fused into a single piece. The tiniest mites are wormlike and suck plant juices, thereby causing galls, spots, and blemishes on the foliage of trees and plants. Other pygmies live in the tracheal tubes of bees and in the hair follicles of mammals, in- cluding man. Some of the most gaily colored species have taken to living in water and swim with the aid of long hairs on their legs. The free living forms abound in detritus, where they prey on tiny animals or eat decaying animals or vegetable matter. About half the mites are parasitic and live on the bodies of animals all or a part of their lives. Mites hatch from eggs as six-legged "larvae," an unusual physi- cal phase for which we still have no adequate explanation. After a period of feeding, the larvae change into eight-legged nymphs, which undergo one or more nymphal stages before becoming the sexually mature adults. Most pestiferous of all are the larvae of the harvest mites, known to Americans as redbugs and chiggers, which attach to the skin and cause violent itching and irritation. Some redbugs transmit Rickett- sial organisms, which cause tsutsugamushi disease, or scrub typhus, which is frequently fatal to man. The nymphs and adults of the redbugs are innocuous creatures content to live on vegetable matter. Largest of all the mites are the ticks, whose leathery bodies are capable of becoming greatly distended with blood, to nearly an inch long in some females. Following engorgement, which is ac- complished by forcing the beaklike mouth parts deep into the skin of the host, the mature females fall to the ground and lay several thousands of eggs. From them hatch six-legged larvae, called "seed ticks," which climb on the body of a new host when opportunity arrives. Some ticks use the same host during all their feeding, but others require two or even three different kinds of hosts in order to complete their life cycle. Many ticks attack man and are a great source of annoyance because of their irritating bite. Among tick- THE PLACE OF SPIDERS IN NATURE 19 borne diseases are Texas Fever of cattle and Rocky Mountain Spotted Fever, a serious illness of man. Solpugids. The curious arachnids know as solpugids (Plate 4 and Plate VIII) are commonly encountered in the American South- west, as well as in some of the northern states in the West. The outstanding characteristic of these creatures is the great size of their chelicerae, which are proportionately larger than in any of their relatives. While feeding on their insect prey, they work the cheli- cerae with a sawing motion, holding fast with one while they drive the other in deeper. It is believed that they take only liquid food from their prey and cannot eat pieces of any size. Most species live in arid regions, where they hide under stones and debris, and come out at night to do their hunting. They are swift creatures and for that reason have been called "wind scor- pions" in the Near East and in Africa, where a great many large species abound. Most American species are about an inch long, but two or three are nearly double that size; with their long legs clothed with reddish hairs, they have a formidable appearance. But, since they possess no poison glands and cannot effectively use their tre- mendous chelicerae on large objects, they need not be feared by man. Ricinuleids. The curious, enigmatic arachnids of this group are the rarest of all arthropods. They resemble ticks superficially in general appearance, and further simulate the sluggish, deliberate movements of the latter. The ricinuleids possess various peculiarities of structure that set them apart from all other living arachnids, and represent a group that was probably far more abundant in Car- boniferous times than they are today. No true eyes are present in the ricinuleids, but vague, pale spots on each side of the carapace may well represent vestigial eyes. Appended to the frontal edge of the carapace is a hood, the cucullus, which fits down tightly over the chelicerae. The cephalothorax is narrowly joined to the abdomen by a pedicel, but this is hidden from view by expansions of the base of the abdomen, which fits very closely with the cephalothorax, the juncture forming a coupling device. The living animal is able to disengage the carapace from the abdomen so that the genital orifice is exposed, and this action is necessary during egg-laying and mating. In the males, the third leg is provided with a complicated copulatory apparatus. It is presumed by analogy that the apparatus 20 AMERICAN SPIDERS aids in the transfer of the spermatophore to the female during mating. However, the exact use of this unique structure has never been observed. A single species of this curious rare group is known from southern Texas. A few other species occur in tropical America and in Africa, but the appearance of even a single example of this order is an event. Whip Scorpions. The whip scorpions resemble the true scorpions in a general way, but are readily distinguished by the absence of a caudal sting and by important differences in the other appendages. The pedipalpi are enlarged into formidable grasping organs, which bear, along their inner edges, numerous teeth or sharp spines that aid in crushing prey. The long, slender first pair of legs is special- ized as organs of touch. The tailed whip scorpions have a slender, jointed, whiplike tail, which is responsible for their common name. In this group the carapace is longer than it is broad, the pedipalpi are very stout, and the first pair of legs is of only moderate length. Essentially noc- turnal in habit, these creatures spend the day in crevices in trees or under objects on the ground, and are active burro wers into sand and debris. Although greatly feared by uninformed peoples, the whip scorpions are without poison glands and incapable of causing more than slight mechanical injury with their clumsy, raptorial pedipalps. At least some of them are known to emit an odor re- sembling' acetic acid from glands located in the base of the tail, a fact that finds expression in the name of "vinegaroon" given by some Americans to Mastigoproctus giganteus (Plate VIII), the giant whip scorpion, which often measures three inches long. The tailless whip scorpions are flattened creatures, which again have the carapace longer than broad, but are without any trace of a tail. The first pair of legs is modified into very long, lashlike whips, the tips of which are flexible. These animals live in dark, sheltered places, such as fissures in the rocks and under the bark of trees. They frequently occur in great numbers in caves, and many of them enter houses. They run with great speed when disturbed. Two or three species occur in the southern part of the United States. Micro-Whip Scorpions. As their common name suggests, these tiny arachnids resemble the tailed whip scorpions, but they are far PLATE I Walker. Van Riper Banded Argiope, Argiope trifasciata, swathing a grasshopper Orange Argiope, Argiope aurantia, with swathed prey THE PLACE OF SPIDERS IN NATURE 21 more generalized in thek structure. The largest examples so far discovered are only about one tenth of an inch long, and half of this length is made up of the slender tail. The micro-whip scorpions have no eyes, and their mouth parts are extremely simple. All ap- pendages are leglike and none have become specialized for grasping, cutting the prey, or otherwise aiding in feeding. These minute arachnids are found in Texas and California and in warm areas in other parts of the world. They live under stones and probably feed on tiny insects. THE STRUCTURE OF SPIDERS To understand more fully the accomplishments and limitations of spiders, it is essential to have a brief resume of their most obvious physical features (Text Fig. i). In common with most Arachnida, they have the body divided into two principal regions, the cephalo- thorax and the abdomen, and each of the sections is provided with certain types of appendages. In spiders the division between these two units is a very narrow pedicel; whereas in such relatives as the scorpions, ticks, and mites the waist is thick. From the several nar- row-waisted arachnids the spiders are immediately differentiated by their possession of ventral spinning organs, or spinnerets, on an ab- domen that is completely unsegmented, except in rare instances. Furthermore, it can be noted that the males of all spiders have a complicated copulatory organ on the end of the pedipalp, a structure never found in this position in the other arachnids. Cephalothorax. As the name implies, the cephalothorax repre- sents those segments commonly called head and thorax, but they are intimately fused into a single piece. It must be remembered that several distinct segments have formed this region; their number is indicated by the number of pairs of appendages (in spiders only six) and sometimes by vague indications. The dorsal part of the cephalothorax is provided with a hardened shield or carapace, ordi- narily convex and bearing the eyes at the front end. The head portion is usually more elevated, and may be strongly marked off by a V-shaped groove. On the rounded, flatter thoracic portion are usually evident a median groove and radiating depressions that mark the internal attachments of the muscles of the stomach and of the legs. 22 AMERICAN SPIDERS The cephalothorax is subject to considerable variation in shape and armature. In long spiders it is usually long, and in short species may be wider than its length. Various spines, humps, and promi- nences of many kinds often surmount it; frequently some of the eyes sit on weirdly designed elevations. In the dwarf spiders the carapace of certain males is grotesquely formed, and has deep pits into which the chelicerae of the females are fitted during copula- tion. In most instances the reason for the presence of such special- ized innovations is not clear. On the front of the head are the eyes, which are simple and resemble the ocelli of insects. Most spiders have eight eyes, appar- ently the original number, but various lines have lost some, so that there are in existence six-eyed, four-eyed, and two-eyed spiders. In one tiny spider from the jungle floor of Panama only a single median eye is present, probably representing the fusion of one pair. Some of the cave spiders and others that live in dark situations have completely lost their eyes, or retain only vestiges. The size and position of the eyes vary considerably. Some of the hunting spiders have large eyes and relatively keen vision, this being one of the necessities for their foraging activities. In many, a tapetum, which causes the eyes to shine in the dark when struck by light rays, con- tributes to the efficiency of this night vision. Most spiders, how- ever, are shortsighted animals that rely on their sense of touch, which they have sharpened at the expense of their eyes. Immediately below the carapace on the ventral surface of the cephalothorax is a median plate, frequently heart-shaped, called the sternum. In front of it is the much smaller lower lip, or labium, which forms the floor of the mouth. Around each side of the sternum are the coxae of the legs and the pedipalpi, which fit snugly against the sternum and lie in the space between it and the carapace. The coxa of the pedipalp in most spiders is fitted with an enlarged, sharp plate, the maxilla or endite, which aids in the breaking of the prey. Directly beneath the cephalothorax at the front end are located the two chelicerae, or jaws, which are the offensive weapons of the spider. It is believed that the chelicerae are derived from the same pair of primitive appendages that became the second antennae in the crustaceans, and this fact illustrates the quite distinct use to which the same generalized appendages are put by a different crea- tures. Each chelicera is composed of two segments, a basal one, which is stout and ordinarily margined by a toothed groove at the THE PLACE OF SPIDERS IN NATURE 23 distal end, and a shorter, movable fang, which lies in the groove when at rest. The sharp fang is the part that is thrust into the prey. Near its end is a tiny opening through which venom flows into the wound. The poison glands, present in all but two small groups of spiders, are associated with the chelicerae, sometimes being entirely contained within the basal segment, but in most true spiders extend- ing farther back into the head as more or less voluminous pouches. All spiders are predaceous, subsist on the body juices of living animals, and only rarely can be duped to accept dead food. The bulk of their food is made up of insects, which are subdued by their venom. Their method of feeding is a most unusual one. The sharp edges of the maxillae and the chelicerae are used to crush and break the fresh body of the prey, which at the same time is bathed with quantities of digestive fluid from the maxillary glands. The softer parts of the animal are broken down and predigested to a liquid state, and this liquid is sucked into the stomach by means of powerful muscles. As the prey is rolled and chewed, it gradually becomes smaller and smaller until only a little ball of indigestible matter remains. This is finally cast aside, or, in some instances, is hung up on the egg sac or in some section of the web, a trophy of the chase. In some hard-bodied insects the juices are sucked through holes made by the chelicerae, and the shell of the drained insect is then discarded. Some spiders require several hours of nearly continuous effort to digest completely an ordinary fly. It is doubt- ful that spiders ever actually imbibe solid food material through the small mouth, and probable that even small snakes, birds, and mam- mal prey are first reduced by the powerful digestive juices. The remaining appendages of the cephalothorax are the pair of pedipalpi and the four pairs of walking legs. The former are situ- ated on each side of the mouth and resemble the legs closely except for size and for lack of the metatarsal segment. In the female, the pedipalp is a simple appendage terminated ordinarily with a single tarsal claw, but in the male the distal end is the seat of the special copulatory organ of that sex. The role of the palpi in mating will be mentioned later. Four pairs of legs are always present, as in typical arachnids. Each leg consists of seven segments, called— beginning with the one that fits snugly into the sternal space— coxa, trochanter, femur, pat- ella, tibia, metatarsus, and tarsus. At the end of the tarsus are to be found two or three claws. The legs vary tremendously in length 24 AMERICAN SPIDERS among different spiders, some of them being long, fine stilts on which the spider hangs, and others stubby props. With so many walking appendages, the means of synchronizing all of them is of some interest. In order to take a step, the spider moves the first and third leg of one side in conjunction with the second and fourth legs of the other side of the body. The remain- ing legs of both sides go into action while the other series is at rest, and thus the creature advances step by step. The appendages and other parts of the body are usually covered with hairs and spines of different kinds. Some of these lie flat against the integument and serve as a covering blanket. Others are heavier or longer or more erect, and are used in many ways by the spider to perform important functions during the spinning of silk, for the preening of the body, preceding and during the mating, and as aids in capturing and holding the prey. Many of these setae are extremely sensitive to touch and vibration, and some may be re- ceptors for various chemical stimuli. By means of its sensory hairs the spider has a keen knowledge of its surroundings. Abdomen. The juncture between the cephalothorax and the abdomen is made by a narrow waist or pedicel, which represents the first true abdominal segment. In the antlike spiders the pedicel is visible from above as a small tubular connection armed above and below by hard plates, but in most other spiders it is not evident, its presence being largely masked by the overhanging abdomen. Through the tiny channel of the pedicel must pass the several struc- tures essential to maintenance of life in both body parts: the ventral nerve cord, a large artery, part of the midgut, and, frequently, numerous tiny tracheal tubes. Ordinarily the abdomen is a saclike structure without visible segmentation and, though covered by a sclerotized cuticle, is usu- ally much softer than the cephalothorax. In the primitive liphistiids and their relatives, the dorsum of the abdomen is armed with a series of hard transverse plates, or tergites, each set with erect black spines. In a few of the primitive true spiders there are evidences of dorsal segmentation, especially in the spiderlings, but in some well-known cases this segmentation may have been acquired sec- ondarily. The abdomen frequently exhibits on its upper surface a series of small, rounded depressions that mark the internal attachments of muscles. Often brightly painted, and variegated with contrasting THE PLACE OF SPIDERS IN NATURE 25 colors, the abdomen in many groups of spiders is accorded more than its share of elegance and elaboration. In some spiders the dorsum is covered in whole or part by a hard plate, and in others it is armed with curious spines and processes, some of them of great length. The reasons for the possession of such curious structures are no more apparent than are the reasons for those on the cephalo- thorax. Perhaps, because of its many sharp projections, this armor discourages birds from attack. In some of our sedentary spiders the abdomen is drawn out into a long tail, which gives the creature a worm-like appearance. The under side of the abdomen is much like the upper in many spiders, and rarely bears conspicuous prominences. Near the base are usually to be seen the two openings to the book lungs, and between them the genital opening. The copulatory organ of the mature female, the epigynum, is located just in front of the genital opening and takes one of many forms. Farther back may be present a second pair of book lungs, a pair of tracheal spiracles, or, near the spinnerets, a single median spiracle. In most spiders of the northern hemisphere is found the single spiracle. At the tip of the abdomen is the anal tubercle or postabdomen, which has the anal opening at its tip. Both book lungs and tracheae are found in spiders. The open- ing to the former is a rather conspicuous transverse spiracle, and the area of the lung itself is usually evident externally as a paler patch. In all the tarantulas and their allies, and in one small family of true spiders, two pairs of book lungs are present, the front pair near the base of the abdomen at each side of the genital pore, and the hind pair much farther back near the center of the abdomen. The possession of four lungs is usually considered to be a primitive condition, since higher spiders have the posterior pair changed into tracheal tubes. The tracheae always replace the book lungs when the latter are lost, and probably are not new creations at all but only modified and expanded book lungs without the leaves that ramify beyond the original space limits. In most of the true spiders there is a tracheal spiracle just in front of the spinnerets. In a few tiny spiders all the book lungs have been replaced by tracheal tubes. Because in the higher spiders the book lungs have been replaced, at least in part, by tracheae, it can perhaps be concluded that these latter are more efficient respiratory organs. The true spiders are more vigorous creatures of much smaller average size than the four- lunged spiders, and require superior respiratory as well as other 26 AMERICAN SPIDERS equipment to maintain their place in the extremely diversified hab- itats they occupy. The spinning organs of spiders are the spinnerets, fingerlike appendages usually located near the end of the abdomen on the lower surface. They are believed to have been derived from two- branched abdominal appendages of ancient spiders, or their pre- cursors, which were originally put to some other use than that of spinning, perhaps being used as swimming or ambulatory organs. Associated with each of these appendages was a coxal gland in the abdomen that voided its excretory products through a pore on some part of the appendage. From the two pairs of two-branched ap- pendages of the third and fourth abdominal segments have come the four pairs of spinnerets of contemporary spiders. Their devel- opment, modification, and elaboration have gone hand-in-hand with a metamorphosis of the lowly coxal glands into a series of abdominal receptacles for production and storage of distinct types of liquid silk. Originally an excretory product, silk has been put to varied and distinct uses, and it has largely charted the course spiders have followed through their racial history. The spinnerets were originally located much nearer the base of the abdomen than their position in most modern spiders now indi- cates, and there was a considerable open space between them and the anal tubercle. The trend has been toward reduction of the num- ber of abdominal segments, and the simplification of the systems inside the abdomen, as well as the segmentation of the outer integu- ment. As the posterior segments became superfluous and were lost or incorporated into the anal tubercle, the relative position of the spinnerets changed also. Ancestral spiders had a long interval of segmented abdomen between the spinnerets and the anal tubercle. In Liphistius this space has been greatly reduced by partial reduc- tion of the size of the segments. In Atypus and Antrodiaetus the space interval has been still further reduced, and in almost all other spiders the spinnerets are immediately adjacent to the anal tubercle. Only in the most primitive spiders are eight spinnerets still present as fingerlike projections. The liphistiid spiders have re- tained all of the projections, but both the anterior and posterior median spinnerets are greatly reduced in size, and perhaps figure little or not at all in spinning. In Heptathela only six spinnerets are present, and the so-called seventh one is the fused remnant of the posterior median pair, a "colulus" in an advanced stage of obso- lescence. In the other mygalomorph spiders, the anterior median THE PLACE OF SPIDERS IN NATURE 27 pair has been lost, and in only a few are there vestiges of the anterior lateral spinnerets. Thus the four spinnerets of the tarantulas and most of their allies represent the single, small, posterior median pair and the longer, posterior, lateral, segmented pair. The loss of the spinning function seemingly has preceded the degeneration and obliteration of the spinning organs. Most true spiders have retained the eight spinnerets in one form or another, and in only a few instances have they reduced their number below three pairs. In all the cribellate spiders the anterior median pair is still retained as the cribellum, a flat spinning field which is used in conjunction with a comb of hairs on the fourth metatarsus, the calamistrum, to produce characteristic threads. The cribellum probably existed before the anterior median spinnerets had lost their spinning function, and became greatly changed and important because of its special function. Whether the cribellum is a new development from the ancient anterior median spinnerets, or represents the ancestral condition of all true spiders, is still a debatable question. At any rate, in almost all higher spiders a vestige of variable size evidences the former presence of an anterior median pair of spinnerets. In some it is a fingerlike colulus; in others, a pair of flat plates, connate plates, or a single sclerotized plate, all set with covering hairs; and in still others, a tiny point or blister bearing one or two erect setae. In some groups of true spiders the hind spinnerets are greatly reduced in size and become obsolete to a considerable extent, but their former location is marked by some sort of vestige. The ordinary true spider has three pairs of well-developed spin- nerets set closely together in a single group. The anterior pair is two-segmented, and the apical segment is bountifully supplied with many spools and a fewer number of spigots 'on the spinning field. The posterior pair is likewise segmented, most commonly with two but frequently with three or even more segments, and is also well supplied with spinning equipment. Between the latter are the me- dian spinnerets, each of a single segment and ordinarily less well provided with spinning openings. In the sedentary orb weavers and comb-footed spiders, which are the finest spinners, the spinnerets are relatively short, with small apical segments, and are set closely together in a small field. In many other spiders whose spinning is less noted the spinnerets are sometimes long and conspicuous, frequently many-segmented, and arranged in different ways. CHAPTER III The Life of the Spider BALLOONING M UCH OF THE ADVENTURE IN THE life of the spider is crowded into the first few days of freedom when the young spiderlings, having just broken through the egg sac, strike out for themselves in a world completely new to them. It is spring and the warmth of the sun has changed the inertia of earlier life in the egg sac to one of intense activity. Hundreds of brothers and sisters, still closely packed together and indistinguish- able one from the other, move about within the narrow confines. Finally the actions of a few vigorous leaders result in the opening of a small aperture at some point in the sac, and a little body squeezes through it to greet the open air. One by one the tiny creatures emerge through the round opening, until the sac is covered with them. They do not tarry long but climb all over the dried leaves and the stems on which the sac is placed, stringing their threads as they go. Soon we see a tangle of webs (Plate 8), strung on every available support, crisscrossing in all directions, and invad- ing space like a living thing. Many of the spiderlings hang motion- less once they have gained their particular station, but others press on with undiminished activity. Up and up they move, to the tips of the tall grass stems and the summits of the leafless shrubs which mark the meadow site of the egg sac. Straight toward the sun they climb until they can climb no higher, impelled by a strange urge to throw silken threads out upon the soft breezes. This is the urge toward ballooning, one of the most extraordi- nary accomplishments of the spider. Once the spiderling has reached the summit of the nearest pro- montory, a weed, a spike of grass, or a fence rail, it turns its face in the direction of the wind, extends its legs to their fullest, and 28 PLATE III George Elwood Jenks Female bolas spider, Mastophora cornigera, with recently emerged brood, including some adult males Walker Van Riper A symmetrical orb web of a mountain orb weaver, Aranea aculeata Walker Van Riper Meshed web of Dictyna on dried weed PLATE IV a. Preparing to leap Walker Van Riper b. Leaping A JUMPING SPIDER, Phidippus audax, AND ITS DRAGLINE Walker Van Ripa THE LIFE OF THE SPIDER 29 tilts its abdomen upward (Plate V). The threads from the spin- nerets are seized and drawn out by the air currents. Although the dragline threads are often used, those from several spinnerets may stream out in long filaments. When the pull on the threads is suffi- ciently strong to support the weight of the aeronaut, it lets go of the substratum and is pulled into the air. Spider lings balloon in different ways, and some of them when afloat climb on their threads like little acrobats, pulling in and winding up or streaming out more filaments, and in this way exercising some control of the ship they are flying. Not the exclusive habit of a single species, as was once supposed, or limited to part of any season, ballooning goes on during much of the year and is easy to observe. In the spring and during the fall months, when immense quantities hatch from the egg, emerge from their egg sacs and fly, the ballooning spiders by their very numbers force themselves upon our attention. Small spiders can be inspired to take off if one blows steadily against them; they tilt up their abdomens, assume a ludicrous pose, and then bound into the air. Because they are so tiny and weigh an insignificant amount, spider- lings are sometimes at the mercy of the air currents and are lifted into the air when they least expect it. Even larger spiders, caught while dropping on their threads, are blown some distance. The small aeronauts seem to float on streamers only a yard or two in length, but the lines may actually be several times as long. In the days of Aristotle, it was commonly believed that the spider could shoot out its silk as the porcupine does its quills. We know now that the spider must depend on breezes to pull the threads from its spinnerets and to bear it aloft after the volume of silk is great enough to support its weight on the air currents. How far do spiders fly on their silken filaments? Darwin recorded the arrival on the Beagle of "vast numbers of a small spider, about one tenth inch in length, and of a dusky red color," when the ship was sixty miles from the coast of South America. He watched them and observed that the slightest breeze was sufficient to prompt them to sail rapidly away, after letting out new lines to catch the wind. Even greater distances have been covered by these tiny aeronauts, which have been known to alight upon the rigging of ships more than two hundred miles from the nearest land. Because they move upward and forward at a substantial pace, and because of their tiny size, the spiderlings are quickly lost to sight. The average distance they span can only be conjectured. The spider may be dropped to 3o AMERICAN SPIDERS the earth near the site of its departure, but it can fly again and again, and thus accumulate a substantial dispersal distance. Most ballooning goes on at reasonable heights, probably less than two hundred feet, as was noted by McCook; but sometimes powerful air currents carry the creatures to great heights. During an aerial survey in Louisiana, B. R. Goad found spiders and mites well represented in samples of aerial fauna at 10,000 feet, and they were even more frequent in the catches of from 20 feet up to 5000. Ballooning has made possible the distribution of spider species over the world. Species have been enabled to send pioneers in num- bers into new areas of all kinds. Oceanic islands have received their spider population almost exclusively through this colonizing mech- anism. On the bleak cliffs of Mt. Everest, at an elevation of 22,000 feet, Kingston found tiny jumping spiders hopping about on the surface and hiding underneath stones. These could easily have been carried upward by the air current. On the other hand there is a pos- sibility that they were permanent residents living at an elevation too high for almost any other creature, and undoubtedly existing on small insects unnoticed by Hingston. In the temperate zone aeronautic spiders are most numerous during Indian summer, when balmy days follow cool nights. In 1918, J. H. Emerton studied the aerial fauna in Massachusetts and listed sixty-nine species that took to the air during the days of his observation. A considerable number of these spiders were fully mature, others were advanced in their age, but all were of rather small species. It is now well known that many adult and half-grown spiders fly, and that this curious activity is not confined to spider- lings just emerging from their egg sacs. Emerton characterized the males of Zygoballus terrestris, a stocky little jumping spider, as be- ing "a regular autumn flyer." Males of some of the smaller orb weavers, such as Aranea pegnia and A. displicata, may be seen ballooning on sunny afternoons, floating a few feet above the ground on long filaments. It is probable that almost all groups of true spiders use this interesting dispersal device during at least some part of their life. Those that shun the light during all their life may not resort to flying; and only a few of the mygalomorph spiders are credited with this activity. The tarantulas are not known to balloon at all, and the large size of their young would seemingly preclude such activity. The purse- web spiders, notably the European Atypus piceus, disperse by taking to the air for short distances, so it is prob- THE LIFE OF THE SPIDER 31 able that many of the smaller four-lunged spiders also have this singular habit. A few years ago Dr. W. J. Baerg described the flying activities of one of the trap-door spiders, Pachylomerus cara- bivorus. The young leave the parental burrow and walk in single file toward and up a tree of considerable size, leaving behind them as a record of their march a silken band that can be traced back to the trap door. From the tree the plump little creatures "spin out a thread of silk, which, when having sufficient buoyancy, carries the spiders off and out into the world." Dr. Baerg did not see the babies fly, so we know nothing of the distance they covered or of their flying behavior. However, this activity may be limited to certain species or only indulged in occasionally. The young of some Mexican species of Pachylomerus remain in the burrow with the mother until they are much too large to balloon. The drifting threads of spider silk are known in prose and poetry as gossamer, a name of uncertain derivation but possibly from "goose summer" in "reference to the fanciful resemblance of the fragile skeins of silk to the down of geese, which the thrifty housewife causes to fly when she renovates her feather beds and pillows." The gossamer season is known in France as fils de la Vierge, and in Germany as Marienjaden or "Our Lady's threads." The reference here regards gossamer as being "the remnant of Our Lady's winding sheet which fell away in these lightest fragments as she was assumed into heaven." Great showers of gossamer have fallen in many places in the world, and their origin has been subject to fantastic interpretations. The true explanation is a very simple one. During the autumn months, spiders become greatly active and cover the meadows and shrubbery with innumerable filaments, which soon form a thin web- bing over everything. Many of these threads are put out by spiders in unsuccessful attempts to fly, and remain hanging on the vegeta- tion. The matted gossamer is then picked up by the wind and showered down in spots often far from where the cobweb orig- inated. In the Yosemite valley of California is located a series of arches which form natural traps for spider threads carried upward by the air currents and deposited in vast sheets. In these areas "all the shrubs, bushes and trees are webbed about in such a manner that the trunks of the largest trees are but faint shadows, while limbs and foliage resemble a glistening mass of crystal. In the midst of this mass are bunches of rolled-up webs that are as white as cotton 32 AMERICAN SPIDERS and quite thick. When the mass is disturbed by a gentle breeze, it moves throughout its entire length with a graceful undulating mo- tion." 2 The gossamer that falls during rainstorms in California may well have its origin in some such concentrated area of silk. It is generally believed that ballooning and its resultant dispersal is an instinctive impulse based on necessity, and that it constitutes a protective device. The scattering of the many babies from the site of the egg sac apparently works against overcrowding and fratricide, and improves the chances of survival for each tiny aero- naut. However, we must remember that flying is not the province solely of the spiderling, and that spiders of all ages indulge in it, limited only by size and weight. During their babyhood spiders eat very little and probably represent no great menace to each other. On the other hand, a high percentage of aeronauts may drown or be dropped in situations where they have little chance of survival. THE EGGS The life of the spider begins at the time when a zygote is formed by the uniting of the male spermatozoon with the ovum of the female. It is believed that this occurs soon after the eggs are laid by the female. The mother spider prepares a silken sheet on which the eggs are placed. They issue one by one from the genital opening beneath the base of the abdomen, and are bathed with a syrupy fluid in which quantities of sperm from the stores in the spermathecae have been discharged. At this time the eggs have a very soft chorion, which is easily penetrated by the sperm at any point. Spiders have long been listed among animals that are able to reproduce parthenogenetically, that is, without having the eggs fertilized by the male gamete. This belief has been perpetuated on the basis of a few records, which are now completely discredited. It has become well known that females can store the sperms of males for weeks or months, and that they are thus able to fertilize several masses of eggs in succession at distant time intervals from the product of the initial fertilization. This curious fact has prob- ably misled the few workers who have recorded parthenogenesis in spiders, a phenomenon for which there is no unassailable evidence. 2 L. O. Howard, "On Gossamer Spider's Web," Proc. Ent. Soc. Washington, Vol. 3, pp. 191-2. Walker Van Riper, Colorado Museum of Natural History A humped orb weaver, Aranea gemmoides, on egg sac PLATE 6 Walker Van Riper, Colorado Museum of Natural History a. Black widow, Latrodectus mactans, in web J. M. Hollisler b. Black widow, Latrodectus mactans, ventral view BLACK WIDOWS THE LIFE OF THE SPIDER 35 After laying a mass of eggs, the female covers them with a silken sheet and molds the mass into the egg sac characteristic of her spe- cies. The eggs (Plate X) are ordinarily spherical, or broadly oval, but their shape may be largely determined by their position in the egg mass. A great many spiders cover the eggs with a viscid secre- tion, which hardens and agglutinates the mass into a single body. In some cases the eggs are only lightly agglutinated, held together in a mass by a few threads, and thus retain nearly a spherical form. Frequently, the weight of the mass is so great that the eggs assume the shape forced upon them by the available space, and thus are irregular in outline. The size of the eggs varies within rather wide limits, being 0.4 mm. in some of the smallest true spiders, but at- taining 4.00 mm. in the large tarantulas. The number of eggs laid by different spiders varies enormously. The largest of all spiders, Theraphosa blondi, is reputed to lay as many as 3000, and the large orb weavers and pisaurids, which fre- quently spin more than a single egg sac, are credited with 2200 in a single sac. At the other extreme we find many tiny spiders that habitually lay only one, two, or very few eggs at a time, and per- haps no more than a dozen during their lifetime. The average num- ber for average spiders is in the neighborhood of one hundred. Those habitually producing more than a single sac usually place fewer eggs in each, so that the average is not greatly increased. There is a considerable correlation between the size of spiders and the number of eggs they are physically capable of producing at any one time. We expect the large tarantulas to be large egg pro- ducers, and find it true, as is well shown by Baerg's average of 812 eggs per sac for one of the large southwestern American species. The contents of five sacs varied from 631 to 1018. The sacs of these creatures are tremendous flabby bags often 2 or 3 inches in diameter. An unopened sac of Hapalopus pentaloris, a brightly colored and curiously marked tarantula of moderate size, from Mexico, con- tained 986 young and each of the babies measured about 3 mm. in length. Only 288 eggs were found in a sac of Phormictopus can- ceroides, a very large West Indian tarantula. Another unopened sac of this same species was 2% inches in diameter and contained 252 eggs in the deutovum or second egg stage. The eggs of the first sac measured about 4 mm. in diameter, and the deutova of the second were about 7 mm. in length. Larger eggs are produced by spiders of greater size. The eggs of Phormictopus are as large as small peas and exceed by several 34 AMERICAN SPIDERS times the bulk of those of any true spider. The young of these spiders after the first true molt are quite large, 7 mm., even before they have left the egg sac. It is small wonder that ballooning is not a characteristic of this group of spiders. Some true spiders produce a greater number of eggs during a single year, but female tarantulas live several years, and in total number of eggs produced probably far outdistance all spiders. True spiders may produce few or many eggs and may place them in one or in several separate cocoons. A tiny cave spider, Telema tenella, lays one egg at a time. The blind spider of Mam- moth Cave in Kentucky is said to lay from 2 to 5 eggs. Among the more generalized true spiders those of the family Oonopidae lay few eggs, and Oonops pulcher of Europe is known to place only two in a cocoon. The Peckhams state that Peckhamia picata, a small, antlike spider, produces 3 eggs. They assumed that ants had few enemies— a supposition for which there seems to be good evidence— and that creatures resembling them would not have to produce so many offspring to maintain their normal population. Likewise in many other families, small spiders produce few simply because the abdomen is too small to accommodate many eggs, each of which must provide sufficient food for the growing embryo. They mul- tiply their low production by maturing eggs for several distinct layings. Medium-sized spiders produce moderate numbers of eggs. Tra- chelas tranquillus, a common eastern American species often found in houses, lays 30 or 40. Many small wolf spiders produce 100 or even less. The common labyrinth spider, Metepeira labyrinthea, spins 5 or 6 cocoons and places about 30 eggs in each. Uloborus americanus also places a string of cocoons in her orb web and leaves about 50 eggs in each. A species from the high mountains of Ari- zona, Uloborus arizonicuSj is a social spider and spins several sacs in each of which are about 60 eggs. And finally, the gregarious Uloborus republicanus of the American tropics, somewhat larger in size than the other two species, spins larger cocoons, in which are as many as 163 eggs. The eggs of these three species are essen- tially the same in size, measuring from .6 to .7 mm. The large orb weavers produce several hundred eggs. The Peckhams state that the orange garden spider lays from 500 to 2200 in its cocoon, but McCook believed that 1000 was about the aver- age number for the species. The cocoon of one of the large fisher spiders from Oklahoma, Dolomedes triton, contained 1537 eggs in THE LIFE OF THE SPIDER 35 its large brown egg sac. The smaller Pisaurina mira had 518 in a sac of average size. Bonnet records a total of 2292 eggs in the four cocoons of the European Dolomedes fimbriatus, a species much smaller in size than several American members of the genus. When multiple cocoons are spun by a single female, the number of eggs is less in the later ones. A female of Aranea cornuta, which made 10 sacs, laid a total of 1210 eggs, deposited in the following order: 234, 218, 182, 140, 112, 87, 81, 72, 51, and 33. In instances of this kind, some of the later eggs may be infertile, owing no doubt to the exhaustion of the semen stored in the receptacles, and perhaps also to its gradual loss of viability. In the later cocoons, the exhaus- tion of the female is apparent in her spinning ability, which becomes progressively less perfect. In order to maintain the normal popula- tion of a species, spiders produce a sufficient number of eggs to cope with all the factors of the environmental resistance, and emerge with a pair for each one in the normal population. The female Argiope aurantia lays 1000 eggs, covers them with a tough cocoon, and yet has an average survival from the large number of only one pair. Peckhamia picata lays a few eggs, placing them at different places in 3 or 4 cocoons, and still maintains an average population. THE EGG SACS The essential work of the female is completed when she has laid her eggs and enclosed them in some kind of silken sac. This act frequently represents the last effort of the mother in behalf of a new generation she may never see. But though early death is the lot for many, it may be delayed long enough for the mother to guard the cocoon for a limited period and even to aid in some way the emergence of her brood. In some species, the female spins more than one sac and must then dispose of the others in her web or hide them away, in order to give her time to the newest sac. In the contents of the sac rest the hopes of the whole species for survival, so it is not surprising that considerable attention may be given to the fabrication of the covering. Many egg sacs are strongly made, beautifully designed creations, often pleasingly tinted with colored silk. Especially constructed for her eggs by the female spider, the egg case is fundamentally different from an insect co- coon, which is the covering the larval insect spins around itself and in which it transforms. The degree of perfection of the sac is cor- 36 AMERICAN SPIDERS related to some extent with the danger of destruction to which it is subjected. When the mother spider remains with her eggs until the young hatch, the need for a tough sac is not so great. Similarly, a sac hidden away in the depths of a burrow or surrounded by barriers of dry web or viscid strands is usually not strongly made. The situation in which the sac is placed and the length of time it must remain there before the young desert it are the important con- siderations. Probably in response to such stimuli, spiders have devel- oped different means of achieving a normal hatching of progeny under varied circumstances. Most spiders are provided with a set of glands especially used for the building of egg sacs. Known as cylindrical glands because of their form, they feed their products through spigots on the out- side of the posterior spinnerets. The silk spun from these glands is frequently different in color from the dry silk, and from that pro- duced by other glands. In addition, the silk of the egg sac is differ- ent in its physical properties, being less elastic and not as strong as the dragline silk. It is apparently never viscid. The outer, varnished layers of some sacs suggest that the outer envelope is different in origin from the silk of most of the sac, or differs at least in the manner of being carded and applied as a layer. The egg sac is generally a spherical or lenticular object, resem- bling a little ball, a biscuit, or a flat disc. The manner in which these sacs are produced illustrates the fact that even in realizing such commonplace structures, the spider must give considerable time and exercise great instinctive ingenuity. Take for example the small wolf spiders, whose sac-making can be conveniently observed. Or- dinarily, Pardosa spins a light scaffolding of lines attached to adja- cent objects, and between them lays down a flat sheet of silk. This sheet usually takes the form of a circular disc approximating in diameter the length of the female. It is a closely woven fabric made by brushing the hind spinnerets from side to side and rotating the abdomen and body. The finished base may be nearly flat, but fre- quently it is a shallow basin, a veritable cradle for the eggs. The actual deposition of the yellowish eggs requires only a minute or two. The gravid female stands over the sheet and ex- trudes through the oviduct a viscid fluid that forms a pool on the silk into which the eggs, singly or in small groups, are laid. The vis- cosity of the fluid is such that the egg mass largely retains its globular shape. In this fluid are sperms from the seminal recepta- cles. The female next spins, over the mass, a somewhat smaller THE LIFE OF THE SPIDER 37 covering similar in texture to the base, and then cuts the biscuit- shaped object loose from the floor and the scaffolding. This she now seizes and holds beneath her cephalothorax and revolves slowly by means of her palpi and legs. At first the spinnerets sew up the edges between the two circular sheets until the break is scarcely apparent. Then the mass is revolved in all directions and the spin- nerets put down additional layers of silk until, as the sac is molded and shaped, a nearly spherical object results. Soon after completion of the sac, its white silk takes on a tinge varying from gray to yellow, blue, or green; and the spider attaches the bag to her spin- nerets. Many spiders spin this type of sac. The great flabby egg purses of the tarantulas are prepared in the burrow and are guarded by the mother until long after the young emerge. The delicate silken bags of the trap-door spiders often hang from the side of their burrow. The large lens-shaped bag of the huntsman spider is held beneath the body by the female, who will not relinquish it without a struggle. Many of the vagrant gnaphosids guard their eggs, but others place their tough little sacs— colored a shiny yellow, pink, or red— close against a rock or a chip of wood and leave them. The simplest type of egg sacs are those of the long-legged pholcids and other primitive spiders, which use only a few threads of silk to hold the mass together. The cosmopolitan Pholcus (Plate XIX) glues her few eggs lightly and carries the mass in her cheli- cerae. The tiny funnel-web tarantulas of the genus Microhexura also carry their eggs in this manner, and thus minimize the need for a strong sac. Some of the most marvelous and elaborate egg sacs are spun by the sedentary spiders, which put their web-spinning superiority to good use in constructing the coziest of egg cradles. The sac may hang in plain view among the threads— the central theme of the web— or it may be tied nearby to herbs or similar objects. Along with the special attention accorded the precious egg mass goes a somewhat different method of realizing the finished cradle. For example: The large, pear-shaped sac of the orange Argiope (Plate 7), which hangs near her web, is constructed in a most unusual manner. Argiope always hangs downward from the threads of her slightly inclined web, and her spinning activities are profoundly influenced by this posture. A series of cross lines attached at several points prepares a firm scaffold for the sac, which itself is a compound 38 AMERICAN SPIDERS structure. First, yellowish threads are laid down to form a roughly rectangular roof, and on this the female spins a thick tuft of fluffy yellowish silk, which forms an irregular mass above her. Into this yielding feather bed she next spins a firmer sheet of dark brown silk, comparable to the base in which Pardosa places her eggs, and which serves the same purpose for Argiope. She lays the eggs up- ward against this brownish sheet by forcing the viscid liquid and the many hundreds of eggs through the genital orifice. (Most of the sedentary spiders that hang downward from webs, and even some of the vagrants that run upright, defy gravity by depositing their eggs in this strange manner.) The egg mass hangs as a yellow spherical ball, and over it Argiope spins a thin but tough covering of whitish or yellowish silk, which is joined to the brown silk disc. Around the whole mass— the eggs, their covering, and the rectan- gular roof— she then spins a fluffy covering of rusty brown or yel- lowish brown silk, very loosely packed, which forms a voluminous blanket around the egg mass. These lines are spun with the aid of the spider's hind legs, which comb them out of the hind spin- nerets in loose loops and pat them down into the mass. Over the spongy padding Argiope now puts down a more finely spun cover- ing of white or yellow silk, largely made by using the hind spin- nerets alone. Smooth and closely spun, this outer covering hardens, becoming a dry yellowish or brownish cover that crackles like parchment. The orange Argiope thus produces, after several hours of tireless spinning, six different sheets, tufts, or covers and from them makes three envelopes for her eggs— a thin white inner fabric, a thick woolly or flossy blanket, and a tough outer cover. The innermost layer is essentially the same as that spun by Pardosa and many other spiders, and is composed of two parts, the sheet that receives the egg mass and the cover. In some orb weavers, the sac is drawn out into a short or long neck or stalk. Mastophora hangs her sac (Plates III and XXIII), a globular bag with a thick stalk once or twice its length, on twigs and leaves near her nest. It is doubtful that the stalk contributes in any way to the security of the eggs, since the sac is easily available to any insects that can reach the twigs. In many instances, Masto- phora lashes the base of her sac directly to the twig. In some other spiders, however, the ball of eggs is suspended in midair by a thread of silk. The pale brown bag of Ero, with its irregular covering of brownish silk, hangs on an inch-long pedicel in a cavity beneath THE LIFE OF THE SPIDER 39 a stone or under boards. The golden brown balls of Theridiosoma frequently are found hanging to vegetation, suspended by a fine long thread. Very likely such a pendant sac offers difficulties to predators that might destroy it if it were nearer at hand. The use of silk coverings to give the eggs a relative security from depredation must have been discovered early in the history of spiders. Even a superficial silken covering would be a deterrent, since many insects cannot penetrate it and might even become en- tangled in the threads. Spiders have, in the course of time, added many refinements to their sacs and thus gained greater protection from predators. The covering has been toughened, thickened, vari- egated with tufted and woolly silks, and, in many cases, several blankets envelop the egg mass. Often the sac is plastered with layers of mud, or embellished with bits of wood, leaves, stones, and other debris, rendering it less conspicuous. Some are glued to stones, tied to twigs, enclosed in folded leaves, or suspended at the end of fine threads. Others sit in the center of the web or lie behind a tangle of threads in a retreat. Some spiders have divided the risk by putting their eggs in several baskets. They spin a series of sacs, which hang as a string in the center of their snare or are left singly here and there. It is uncommon to find every sac in a string parasitized, whereas the whole effort of a mother spider may be lost in a single bag. In addition to this type of protection, the spider often plays an active role in seeing her eggs through to hatching and babyhood. The crab spiders and many hunting spiders guard the egg sac and strenuously resist effort to pilfer the contents. Wolf spiders drag their sac attached to their spinnerets, and, later, carry the young around on their backs until the spiderlings are able to fend for themselves. The varied efforts made by mother spiders to provide for the welfare of their eggs or young are remarkable and complex, and especially noteworthy because they are largely instinctive activities. HATCHING AND EARLY DEVELOPMENT Spiders undergo a development within the egg that is compara- ble to that of other arachnids and also of insects. The embryo spider gradually takes form on the outside of the vast sphere of yolk that makes up most of the egg. On the generalized part, which will become the cephalothorax, appear little buds, which gradually 40 AMERICAN SPIDERS become differentiated into the chelicerae, palpi, and the legs. A similar series appears on the abdominal portion, associated with a rather definite segmentation of eight to twelve segments, but all those behind the sixth true segment disappear as development pro- ceeds. The basal pairs of buds persist for some time, and those of the fourth and fifth segments develop into the paired spinnerets. The buds on the second and third segments become invaginated and go to form the book lungs. Finally, the embryo nearly encircles the outside of the egg and the ventral surface is outside, unnaturally bent and convex so it can lie within the stiff chorion. At this point occurs what is called "reversion," a process by which the position is reversed and the cephalothoracic portion becomes free. At about this time too, with the pressure against the chorion of the expanding embryo, and with the aid of a sharp egg tooth at the base of the pedipalpi, the egg covering is broken. With the shedding of the chorion of the egg there is revealed a creature somewhat spiderlike and yet obviously different from the well-known spiderling. (The term "larva" has been applied to this stage, but since that term more commonly describes insects at an active feeding stage and has quite a different sense, it will not be used here.) It is not unreasonable to suppose that this imperfect creature is prematurely hatched, and that it actually represents part of the egg stage. In order to gain space for fuller development and more freedom, the tough chorion is broken but the creature is still swathed in embryonal membranes. In mites an analogous stage is called the deutovum, and is so similar to what exists in spiders that the term may be applied to the latter also. This period in the spider's growth is not nearly so simple as was once supposed. Dr. Ake Holm has discovered and described in various Swedish spiders two or even more incomplete stadia (the intervals between molts), each marked by the shedding of a mem- brane. Some spiders hatch from the egg at a more advanced stage than do others, the degree of development being roughly approxi- mated by the specialization of the family. In Segestria, the first postembryonal stadium brings to light a very primitive creature, whereas in a more highly developed spider, such as Pardosa, the deutovum is far more advanced. The deutovum is without dark coloration of any kind, the cara- pace usually being milky white and the abdomen somewhat duller. Tarsal claws are completely lacking on the pudgy legs. The crea- ture is unable to feed or spin, for only parts of the important struc- THE LIFE OF THE SPIDER 41 tures are developed. No setae or hairs are present on any part of the body. The shape and size of the eyes are sometimes indi- cated even at this stage, but they are colorless and without func- tion. In the abdomen is an abundant yolk material on which the creature can subsist until able to feed. The deutovum grows quickly after emerging from the egg covering, and soon is twice as large as the space occupied by the egg. The duration of the deutovum stage is usually quite short, and toward the end of it we begin to see the darker coloration of the growing spiderling beneath the cuticle. The first true molt, always undergone while in the egg sac, brings to light the creature that we all recognize as a spider, and which is truly a miniature of the adult. During a rather indefinite period of its life, perhaps for several stadia, it is referred to as a spiderling because of its small size. The legs are now longer, much more slender, and clothed with darker spines and hairs. At the tip of the tarsi are found tarsal claws, two or three depending on the family to which the spiderling belongs. The spiderling is now able to spin but it uses little silk until after it leaves the egg sac. The digestive system is more perfectly developed, and the spiderling is probably able to feed, but its food requirements are still being met by unused yolk material in the abdomen. What happens next is largely dependent upon the tempera- ture. If the weather is favorable, the spiderlings become active and move about in the sac, their actions dependent upon the de- gree of warmth that penetrates through the silken covering of their domicile. Some female spiders guard the egg sac until they die, and others are reputed to aid their babies to escape from the sac by tearing it open. In most cases, however, the female has long since died and the escape must be effected by the spiderlings themselves. In tough sacs they usually cut a neat round hole, through which they emerge one by one, or, in weaker sacs, they will force a large rent. Following emergence comes the dispersal of the family, usu- ally by ballooning. If the weather is cold, the spiderlings in the cocoon are inac- tive. They often stay in the egg sac through the whole winter, awaiting the proper temperature in the spring before dispersing. This is particularly true of those species that lay their eggs late in the fall, when not enough time and warmth are available to allow the spiderlings to develop and disperse. 42 AMERICAN SPIDERS MOLTING At rather definite intervals in its development the spider casts off the bonds of its stiff outer covering and readjusts itself for life in a more advanced stadium. This molting, or ecdysis, is characteristic of all the arthropods and is ordinarily their method of providing for increase of size when the old cuticle becomes too tight. They emerge from their transformation with shiny new armor, fully set with new hairs and spines, and often even with new structures not represented in their previous condition. In the spiders, metamorphosis brings with it a rather gradual change from the spiderling to the adult, and is comparable for the most part to the changes undergone by grasshoppers and other lower insects. During each molt the epidermis formed under the old cuticle is capable of considerable increase in size before it becomes hardened. A much greater change occurs at the last molt, for it brings to light the fully developed, sexually mature adult. Only some of the more primitive spiders resort to molting after sexual maturity; they are the females that are long-lived and perhaps require a periodical change of rai- ment for other reasons. Apparently postnuptial molts are not neces- sary for growth, inasmuch as the creatures have reached their maximum and may even decrease in size thereafter. Perhaps they are required in order to provide a new and complete covering of spines and hairs, which are the prime sensory equipment of spiders, and without which they remain at a distinct disadvantage. Molting is ordinarily preceded by various symptoms that indi- cate the approach of the ordeal. This is particularly true of the later molts, which are of longer duration and more difficult of suc- cessful completion. For hours, days, or even weeks, the spider re- fuses to feed and becomes more and more lethargic. Certain changes in color have been noted, in some instances a darkening of the legs, in others a lightening or darkening of the whole body, owing no doubt to the changes going on under the old integument. Burrow- ing spiders often spin up the entrance of their burrows or block the opening with a plug of earth. Those that normally live in silken nests or leaf retreats use these for molting quarters. Some of the orb weavers hang exposed in their webs and are thus in an especially vulnerable position. The details of molting (Plate X) vary little among groups of spiders, but they are of considerable interest. The large American THE LIFE OF THE SPIDER 43 tarantulas are fine performers and their molting activities have been described a number of times. During the late summer they usually show evidences of an impending change and refuse to accept food for days or even weeks. The dorsum of the abdomen has by this time usually been rubbed completely bare, as a result of the normal scraping characteristic of these creatures; and because they have worn their covering of hairs for a full year, their bodies are dull and quite bleached as compared with their fresh condition. Tarantulas ordinarily have their quarters liberally covered with silk, but on this occasion they spin an expansive, closely woven sheet of silk, appropriately termed the molting bed, which requires several hours of intensive work. On this soft cover the spider lies, turned completely over on its back and with legs outstretched, the front and hind ones with tarsi affixed to the silken bed. To all appearances it is dead, but if one watches the prone figure closely, occasional slight movements can be detected. After two or three hours, the old skin splits along the sides of the carapace, and the old shield comes loose from the new integument. Splitting continues over the pedicel and the sides of the abdomen until the dorsum of the whole spider is partially freed, the old skin adhering more or less closely for some time. At this stage, the spider has changed its position so that it is lying on one side, and it now begins the labori- ous process of pulling the appendages from their old casings. The spider extracts its chelicerae first, and then starts a series of rhythmi- cal contractions which gradually bring to light the femora, patellae, and successively the rest of the legs. The first legs and the palpi are freed initially, then come the posterior legs. After about an hour, the cephalothorax and the legs are completely freed; where- upon the spider easily extracts the abdomen and moves away from the cast skin. For three or four hours it lies on its back or side while the new skin hardens; then it resumes its normal upright posi- tion. The freshly cast skin of the tarantula is moist inside, and the new cuticle also shows traces of moisture. The molting fluid be- tween the old and the new skins perhaps aids the progress of the molt by loosening the old skin. Essentially the same picture is presented in the molting of true spiders. The sedentary spiders hang in their webs or in their re- treats. Many of the vagrants spin a few threads in a favorable nook and hang downward, their tarsi fixed in the silken lines and their abdomen supported by a thread from the spinnerets. The cuticle splits around the sides of the carapace and around the abdomen. 44 AMERICAN SPIDERS Then the legs are freed in slow stages by the usual rhythmical con- tractions, the front ones coming out first and finally the posterior pair. By the time this is accomplished, the abdomen is virtually freed and the spider is suspended in the air by the thread from its spinnerets. The whole process requires ten or fifteen minutes, the length of time apparently being governed by the size of the spider. Young spiderlings suspend themselves, molt, and lengthen their legs in less than half an hour, the molting itself often taking only three or four minutes. Half-grown spiderlings require about an hour, and young males and females in the last molt about two hours. Molting may proceed during either day or night, and seems not to be lim- ited by time as are many other activities of spiders. The freshly molted spider is much paler and softer than in the previous instar, and only gradually hardens and darkens its new integument. During this relatively brief period occurs all the in- crease in size of the carapace and appendages until the next molt. Size increase is usually progressive and is determined by the instar and the sex of the spider. In some species the legs increase tre- mendously in length between the instars. Growth commences as the legs are pulled out of the old integument— much as fingers are pulled from a glove. It continues during the time the spider is free of the cast skin and hangs suspended. The appendages are bent back and forth in a regular ritual. Pierre Bonnet has demonstrated with some very ingenious experiments the necessity of these calisthenics following the molting. Without such bending movements the ap- pendages become sclerotized even at the joints and remain stiff. The number of molts necessary to attain maturity varies widely in spiders. Bonnet has shown rather conclusively that size is the de- ciding factor in most species. Tiny species molt few times, whereas large ones molt a greater number of times. Bonnet noted that small species of 5 or 6 mm. in length (Pholcus phalangioides, Uloborus plumipes, etc.) molted four or five times. Species of medium size, measuring about 8 to 1 1 mm. (Aranea diadema, Pirata piraticus, etc.), molted seven or eight times. The larger spiders, 15 to 30 mm. in length, molt ten to thirteen times (Dolomedes plantarius, Nephila madagascariensiSy etc.). The largest of all spiders, the tarantulas, molt even more often: according to Dr. Baerg, twenty-two times for the male of Eurypelma californica. Further, in these spiders various postnuptial molts make it probable that the females molt between thirty or forty times before they die. At the lower limits, only four molts are credited by Bonnet to the male of Nephila THE LIFE OF THE SPIDER 45 madagascariensis; and I have discovered that the male of Mastophora cornigera undergoes only two molts before becoming mature. Even within the same species there is variation in the number of molts. Bonnet found that to become mature, females of Dolomedes plantarius molted as few as nine or as many as thirteen times. The number of molts was to some extent correlated with size, the larger examples requiring more molts, but various other factors were im- portant, the amount of nourishment being one. The males of this same species became adults after nine, ten, or eleven molts, a number similar in the lower limits to that of the female that they resembled in size. In the northern United States the egg sacs of the species of Mastophora are broken open early in the spring and the young disperse. The males emerge either in the penultimate stadium or fully mature, in the latter case having molted only twice. The fe- males are of the same size, and presumably have likewise undergone one or two molts, but they must molt seven or eight times before they are sexually adult. The difference between molts is probably five or six, and reflects an enormous disparity between the size of the sexes, a difference greater than in any other spiders known to me. Time is also important, and whenever maturity is reached quickly for the species, the molts are near the minimum for the species. When maturity is retarded for some reason, more molts are undergone. Abundant food diminishes the number of molts, whereas starving increases the number. For very few North American spiders is the number of molts known. The black widow, Latrodectus mactans, has been studied rather carefully by several investigators, and we find the usual con- siderable differences between the sexes as regards molting behavior. The males become adult after the fifth, sixth, or seventh molt, whereas females are adult after the seventh, eighth, or ninth molt. In the related spider Teutana grossa the males become adult at the sixth or seventh, the females at the seventh or eighth molt. This number of molts is about average for spiders of this general size, and the males almost invariably molt at least one less time than the fe- male. In 1927 Gabritschevsky recorded the time intervals between molts for Misumena vatia as a part of his paper on the change in pigmentation of that species. The synopsis is as follows: Deposition of eggs, July 28; hatching, August 8 (my estimate); first molt, about August 12; emergence from the egg sac, August 14; second molt, August 24; third molt, September 5; fourth molt, September 23; 46 AMERICAN SPIDERS fifth molt, October 17; sixth molt, January 5; final molt, a time after January 5 that was not indicated. Various morphological changes accompany molting; some of them being very significant. The presence of a third claw on the tarsi of very young spiders that are two-clawed as adults indicates that the three-clawed condition is the primitive one. Young wolf spiders have the eye formula of the Pisauridae, a fact which cor- roborates our belief that the former were derived from an ancestor very much like recent pisaurids. The young of Tibellus oblongus, a greatly elongated species, have the general body form and the eye relations of species of the more conservative Thanatus. Each molt represents a crisis in the life of the spider, and brings with it dangers of many kinds. During the transformation the spider is completely helpless, trussed up in old worn clothing and exposed to attack from many enemies. Crickets, sowbugs, meal- worms, and other omnivorous animals, not serious adversaries under normal conditions, are liable to nibble and kill it; it lies vulnerable to attack from the meanest foe. Normal enemies find it completely unable to fight back. Furthermore, the mechanical difficulty of extracting its appendages may prove insurmountable, and the im- prisoned creature will perish, or so mutilate its legs that its chance for life in a hostile world is much diminished. G. and E. Deevey found that nearly half the deaths before maturity (thirteen out of thirty-one) among the black widow spiders they reared were the result of failure to complete a molt. One out of every twelve spiders in the total of one hundred fifty-eight that were studied from hatching to death died from this cause. AUTOTOMY, AUTOPHAGY, AND REGENERATION The spider shares with many other arthropods the ability to drop an appendage without great inconvenience— called "autotomy" —and the ability to replace it in a more or less perfect form by sub- sequent regeneration. This latter power of replacing lost or muti- lated organs is a very old one, and, most strongly developed in lower animals, serves as a device of great importance from the viewpoint of protection and survival. Often the spider is able to escape the clutches of an enemy without greater loss than the shedding of one or two of its appendages. Whereas autotomy oc- curs in spiders of all ages, the regeneration of new appendages is THE LIFE OF THE SPIDER 47 limited to young spiders that have not stopped molting, or to those few primitive spiders which molt after sexual maturity. It is now known that autotomy in the strictest sense— that is, the act of reflex self-mutilation—does not occur in the Arachnida. An appendage is dropped only after a visible effort on the part of the spider, which struggles with such violence that the tension on the member snaps it off at its weakest point. This action was termed "autospasy" by Pieron in 1907; it involves the breaking of the appendage at a predetermined locus of weakness when pulled by an outside force. This locus is between the coxa and the trochanter in the legs in most spiders: a point found by Wood to resist only 7 per cent of the stress that the next weakest juncture, that between the metatarsus and the tarsus, could withstand. In harvestmen, the weakest point is between the trochanter and the femur, and in other animals the break may occur in quite different locations. The reaction of the spider to the loss of appendages varies con- siderably. The loss of one, two, or even three legs in some of the active crab spiders seems to result in little inconvenience to the animal, which runs away without crippling effects. Stocky crab spiders that have lost the first two pairs of legs take up a position in which the short third legs are directed forward as in normal posture, and are able to move about with relative ease. The stout front legs of the crab spiders are at the same time organs of touch and offensive weapons, and when they are lost, the ability to cap- ture flies is seriously impaired. Mature males that have lost some of their long front legs are at a distinct disadvantage during court- ship, and fall easy prey to females not willing to meet them. Autotomy is easy to observe. If a spider is grasped by one of the legs and the animal has a good hold on the substratum, the leg will break loose at the usual locus between coxa and trochanter. On the other hand, if the spider is held in the air and is unable to exert some countering force by grasping an object, it is unable to drop a leg. When held in a pair of forceps, the animal usually twists around, grasps the forceps, and literally pulls its body loose from the leg. The speed with which this is accomplished varies with the species and with the thickness of its appendages, but it is practically instantaneous once the spider begins to effect an escape. If two legs are held firmly, some spiders break both of them easily, but some of the stocky crab spiders are unable to exert enough force to free themselves. Sometimes the spider is seized by a predator that is able only 48 AMERICAN SPIDERS to break the cuticle of the leg, with the result that blood begins to flow through the break. Although the spider may escape other- wise unscathed, this is a most serious situation, inasmuch as an open venous system allows the gradual draining of blood from the body until death occurs. The instinct of the spider is immediately di- rected to a preventive device. The leg is pulled out, and the flow of blood is quickly halted at the normal breaking point between coxa and trochanter. This amputation is accomplished with the help of the remaining legs and the mouth parts. In some instances the spider spins threads and ties the appendage to them, and is able to amputate a whole leg or even a small stump and thus save itself from almost certain death. Autotomy can be put to use by the spider to rid itself of an appendage that is unwelcome for some reason other than injury. The known males of the species of Tidarren have long been ob- served to carry only one palpus, a great bulbous affair held in front of the head. In the antepenultimate stadium, the palpi are only slightly swollen, but after molting the creatures have two tumorous enlargements resembling boxing gloves. So large are these new members that the spider is handicapped by them, and is able to manipulate them only clumsily. The obvious solution to the prob- lem is the amputation of one of these palpi, and this is exactly what the spider does, by a most interesting process. It spins a scaffold of silk, similar to the molting sheet, and, suspended from it by its legs, fixes one of its palpi in the threads. The spider now twists around and around and, aided by pressure from its hind legs, twists off the unwelcome palpus. The spider now has a single palpus, which is held in front of the head and occupies much of the avail- able space. At the next molt it becomes sexually mature, and the vital parts of the palpus are revealed. A second palpus is never regenerated to replace the old one. The spider's instinct to rid itself of an injured or inconveniencing appendage takes precedence over all others, but once autotomy is accomplished, the spider almost invariably does a most curious thing. It picks up the bleeding member and sucks the juices from it, usually discarding it only after it is sucked dry. This autophagy is perhaps as old a habit as autotomy itself, but may not have any especial significance beyond its general interest. Spiders often attack each other, or other prey, and if successful only in securing a leg, will stop and suck it dry in the same manner. The instinct asso- PLATE 7 J. M. Hollisler a. Opened egg sac of orange Argiope, Argiope aurantia J. M. Hollister b. Egg sac of shamrock orb weaver, Aranea trifolium EGG SACS PLATE 8 Cluster of baby orb weavers, Aranea, preparing to disperse THE LIFE OF THE SPIDER 49 ciated with bleeding prey and the taste of the blood prompts the creature. If a leg is lost by an immature spider, it is replaced by a smaller, imperfect replica at the next molt, provided a sufficient time has elapsed between the loss and the molt. The regenerated appendage increases in size with successive molts but never quite attains the full perfection of the normal appendage. The same leg can be regenerated repeatedly, so long as the spider is still immature, but at least three successive molts are necessary to attain a size com- parable to that of the normal appendage. The regeneration of a leg takes a definite course. A good illus- tration is the crab spider, Misumena calycina. The females of this species have white legs, and when one is lost, the appendage that replaces it is shorter, unmarked (as is to be expected), and deficient in the number of spines, as compared with the normal appendages. In the male, however, the first two pairs of legs are banded after the third or fourth molt, and in each successive molt the amount of pigment in the dark annulae increases. If the male loses a leg during the third instar, when it is still white, after the next molt the regenerated leg is wholly white, but the normal front legs continue to increase their annular pigmental areas. After the fourth molt, the regenerated leg becomes annulate, but the depth of the chromation is much less than in the normal leg. In other words, in Misumena calycina a regenerated leg takes on the normal coloration of the leg at the previous instar, and never quite approximates the normal leg in size and color. LONGEVITY Most spiders that inhabit the temperate zones live only one year. The yearly population may be divided very roughly into two faunas, one identified with the spring and the second with the fall. Over-wintered or recently matured males of many vagrant spiders are abundant in early spring and are on hand when their females become mature. The crab spiders are found on the ground, in the corollas of spring flowers, or running over the stems of shrubs and the bark of trees. The grassland teems with jumping spiders, wolf spiders, clubionids, and many other wandering types. The sedentary web spinners are likewise well represented by many species that spin inconspicuous orb webs or tangled webs on the vegetation. In 50 AMERICAN SPIDERS a few weeks most of the males disappear and gravid females are found on all sides, some spinning up domiciles for egg-laying, while others, having already made their first sac, carry it attached to their spinnerets or held in their jaws. By midsummer the possibility of finding males of the spring spiders is not very good, but the females carry on far into the year, often laying several egg masses. Thus, in the spring and early summer we have the males and females of the spring fauna, and the juvenile and growing representatives of the fall fauna. In the fall, the sedentary spiders advertise their presence in a conspicuous manner by great sheet webs and expansive orbs. The males appear in midsummer and early fall and attend the females on the outskirts of their modest webs. In August we find the webs of the grass spiders on grass and shrubs, and can surprise the adults pairing in the funnels. The orb weavers now are attaining maturity in great numbers, and every suitable situation is filled by a web of variable dimensions. Having vastly increased in size, they spin correspondingly larger webs. As the season progresses, the males dwindle rapidly; soon all are gone, having lived the shorter, intenser life identified with their sex. The females lay their eggs and enclose them in sacs of various kinds, which tend to be more substantially built and more heavily insulated than those of the spring spiders. After the killing frosts of November, most adults of the fall fauna are gone, but already the growing young of the spring spiders have attained nearly their full development. During October and Nov- ember, the partially developed spring spiders engage in ballooning activities in company with many precocious fall spiderlings. The fall spiders produce their eggs in the fall, and their young either spend the cold winter months in their cocoons, or, having deserted them, live under debris or in protected places until warmer days allow them to begin their march to full maturity. It must not be thought that the two faunas are discretely sep- arated one from the other. Actually they are bridged by species that mature during the midsummer. Because of multiple cocooning and precocity, or tardiness, of some species, there is a considerable overlapping of the faunas. Further, some species do not seem to conform to any definite pattern, and mature males and females may be found during almost any month of the year. In the American South it is possible to have two full generations during the year. For the most part, however, even in warmer areas we find only one generation each year. Over much of Canada there is only one THE LIFE OF THE SPIDER 51 generation per year, while in the colder northern reaches some of the species probably require two or more years to attain complete maturity. The life of the male is invariably shorter than that of the female. Males of the spring spiders die in early summer after having lived about ten months. The same longevity holds true of the fall species, and the females outlive the males by several weeks. Under laboratory conditions, G. and E. Deevey found that average male black widows matured in about 70 days and lived a total of about 100 days; whereas females matured in 90 days and lived about 271 days. The greatest life span for the males was 160 days, and for the females 550 days. Such data show that mating between brothers and sisters of the egg masses is quite improbable. A number of spiders are known to live more than one year. In the northern United States almost the only ones to do so are the large wolf spiders, which burrow into the soil and probably live several years. Occasionally other spiders live about 1 8 months, such as the large water spiders and the black widows, even though their normal life span is only a year. The more primitive true spiders often live more than a single year. Some of the segestriids and scytodids are said to be peren- nials, and Dr. Lucien Berland kept a female filistatid for ten years. It is probable that all ancestral spiders were longer-lived, and that one of the sacrifices of the modern true spider for the many advan- tages it enjoys is a drastic reduction in life span. It is generally believed that all mygalomorph spiders live several years. The purse-web spiders are reported to live as much as seven years, and the true trap-door spiders are also perennials. Exceeding all other spiders in length of life are the large tarantulas. Dr. Baerg has kept a female tarantula for more than 20 years and believes that 25 or even 30 years probably represents the normal age for females. The males mature in 8 or 9 years, but ordinarily die a few months afterward. CHAPTER IV Silk Spinning and Handiwork SPINNING CHARACTERISTICS HE MAIDEN ARACHNE, DAUGHTER of Idmon of Colophon in Lydia, became widely known for the excellence of her work at the loom. Indeed, her art was so superb that the nymphs from the woods and streams came to gaze upon it. Many wondered whether even Athene, Goddess of Weaving and the Handicrafts, could surpass this maiden, who seemed to have been tutored by the Gods themselves. So confident became Arachne in her amazing skill that she challenged Athene to compete with her. Although affronted by the presumption of the girl, Athene accepted the challenge and wove a tapestry showing the warfare of the Gods and the fate of those who conspire against them. Arachne depicted the love adventure of the Gods with such exceeding per- fection that the Goddess, unwilling to admit that so high a degree of excellence could be attained by a mere mortal, became enraged and destroyed it with a blow from her spinning shuttle. The rash and humiliated Arachne attempted to hang herself, but the noose was loosened and became a cobweb, and the maiden was changed into a spider. Thus disgraced, lying on the rent pieces of her tapestry, Arachne was condemned to perpetual spinning. The Greek word for spider is arachne, commemorating the weaving skill and mythical fate of the imprudent maiden. From it we derive the group name Arachnida, which embraces all the arach- nids or spider like creatures, and also the ordinal names of Araneae or Araneida, exclusively used for spiders. The English word "spider" is a corruption of "spinder," one who spins, and is similar in form to other Teutonic words derived from the same root, such as the Spinne of the Germans. This root per- sists in different form in the words "spinstress" and "spinster," both 52 PLATE V a. Orienting in response to breeze, secured by dragline Walker Van Riper b. Ballooning threads stream from spinnerets Walker Van Riper A JUVENILE JUMPING SPIDER, Phidippus, ON A THIN TOOTHPICK, PREPARES TO FLY PLATE VI a. The cautious approach of the small male Walker Van Riper b. The mating COURTSHIP AND MATING IN THE BLACK WIDOWS, Latrodectus mactans SILK SPINNING AND HANDIWORK 53 having reference to women who spin as a profession, but the latter has acquired a quite different connotation. While most people associate spiders with a silken web of some sort, few are aware of the dependence of these creatures on silk. The ability to spin is an early gift to the spiderling, and is developed after the first molt and before emergence from the egg sac. Imme- diately upon leaving the sac, the spiderling strings out its dragline threads and attaches them at intervals to the substratum. There- after it is never free of this securing band through its whole life, except by an accidental breaking of the cord. The degree of reliance on silk varies considerably among the spiders. The very oldest ones, the precursors of those few we know from Carboniferous rocks, probably had clumsy appendages that were only beginning to be used to comb out a liquid silk. The most primitive of recent spiders are said not to spin a dragline, although they are otherwise probably as well equipped for spin- ning as most spiders, from the evidence of their well-made egg sacs and silken tubes closed with a trap door. The familiar jumping spiders and wolf spiders, so often seen running over the ground or climbing on plants, are vagrant types in which the use of silk is lim- ited. They employ it chiefly for their draglines, for covering their eggs, and for lining their retreats. On the other hand, a vast multi- tude of sedentary spiders are strongly dependent on silk. Some of them have become slaves of elaborate webs and are nearly helpless when not in contact with them. For spiders of this type silk is of paramount importance during the whole life span. The majority of spiders are inveterate spinners and far surpass all other animals in the variety and excellence of their weaving. Some of the other arachnids produce silk, but they use it in a very limited way. The pseudoscorpions have cephalic glands and spin silk through a tiny spinneret located on the tip of the movable fin- ger of the chelicera. Before laying their eggs, these tiny animals build an ingenious little domicile made of small particles cemented together with silk, and lined inside by a covering of silk. A few of the mites also have silk glands and are said to spin threads so fine they are invisible to the naked eye. The so-called "red spiders" are mites of the family Tetranychidae, which cover the leaves of trees with silk and use it as a protecting blanket for their eggs and young. Many insects spin silk and in such profusion that they rival the work of even the sedentary spiders. The unsightly webs of the tent caterpillars are familiar and despised objects to most people but, 54 AMERICAN SPIDERS looked at objectively, they are quite wonderful fabrications. Their tent nests are not far different from some made by gregarious spiders. Many other moths spin silk, but its use is largely restricted to making the cocoon. The most noted insect spinner is the silk- worm, the larva of the moth Bombyx mori, which has been domes- ticated for so long that it cannot now maintain itself in the wild state. It produces cocoons that are easily unwound, and supplies the bulk of commercial silk. The silk of moths, caddis flies, and sawflies is produced in cephalic glands, which pour their contents through a single opening in the lower lip. The threads are usually much thicker than those of spiders. The silk is probably of only one kind. The spider's reliance on silk is well illustrated by the many dif- ferent uses to which it is put. A list of some of these is given be- low, without any attempt at other than a general classification: Protection and Retreats The dragline; the bridge line; the trap line of the orb weavers; the warning threads of Ariadna The ballooning line Attachment discs to anchor the lines The cells and retreats of all spiders Hibernating chambers Molting threads, beds, and chambers Trap-door covers; spinning up of burrows and open retreats Protection of Eggs and Spiderlings The egg sacs The nursery webs of the Pisauridae Web Structures Associated ivith Mating The sperm web of the males The bridal veil of the crab spiders and other vagrants The courtship and mating bowers of the black widow and sed- entary spiders The mating chambers of the vagrant spiders Structures for Stopping and Ensnaring Prey Sheet webs The stopping tangle webs of the grass spiders and the aerial sheet weavers (Linyphiidae) SILK SPINNING AND HANDIWORK 55 The viscid or entangling webs of the orb weavers and certain other spiders The viscid ball and pendulum line of Mastophora, Dichrostichus, and Cladomelea The viscid hackled band in the diverse capturing webs of the cribellate spiders The catching thread of Miagrammopes The retiarius of the Dinopidae Bands for Binding the Prey The swathing band of the orb weavers The swathing film of the comb-footed spiders The swathing band of Hyptiotes The entangling ribbon of the Hersiliidae The capturing band of Drassodes and other vagrant species The above requirements and others not listed are met by the production of different kinds of silk, which are used, seemingly at the will of the spider, either separately or in combination to pro- vide the special threads, desired bands, or drops for a particular project. THE SILK The silk of spiders is a scleroprotein which is produced as a liquid in varied and voluminous abdominal glands. When drawn out of the spinnerets, the liquid ordinarily hardens to form the familiar silken threads. It is believed that the mechanical stretching of the silk during the drawing of the lines is responsible for the hardening, rather than exposure to air or any chemical process. Viscid silk is produced in some of the glands and remains sticky for long periods. An analysis of the silk has shown that it is a com- plex albuminoid protein quite similar to that produced by the silk- worm, although this similarity is denied by some investigators. The silk of the silkworm comes from modified salivary glands located in the head, whereas that of the spider is derived from transformed coxal glands in the abdomen. Spider silk is noted for its strength and elasticity. The tension necessary to bring a compound thread .01 cm. in diameter to the breaking point was once found to be eighty grams. This consid- erable tensile strength, which is said to be second only to fused 56 AMERICAN SPIDERS quartz fibers and far greater than steel, goes hand in hand with great elasticity. The threads will stretch one fifth their length before they break. The strength of the threads is to some extent dependent on the manner in which the spider draws them out, greater speed increas- ing it. When they are drawn speedily, the fibroin chains attain a maximum orientation, which contributes greater strength to the lines. The cocoon silk of the silkworm is essentially equal in strength to that of the orb-weaving spider. However, spiders produce sev- eral varieties of silk, and some differences are found among them in strength and elasticity. The viscid line of the orb-weaver snare is not very strong but extremely elastic; whereas the foundation lines of these webs are of great strength, exceeding even that of the cocoon silk. Most spider threads are not single fibers, although they may appear so to the naked eye. The dragline thread readily separates into two rods of equal thickness, but often elements from other glands lie parallel to these elementary strands and mar the uniform- ity. Under ordinary magnification, single fibers are rather uniform rods, but when photographed by the electron microscope at 35,000 diameters even the finest threads are not completely uniform, and show tiny enlargements and irregularities. Not much detail of the internal structure of the silk can be seen even at this great magnifi- cation. The finest single fibers attain a thinness of 0.03 micron, or about one millionth of an inch, and are invisible to the naked eye. Much thicker threads are relatively large, being o.i micron, or one quarter-millionth of an inch in thickness. Many molecules are larger than the width of these spider threads. It is possible that the spider can draw out its filaments to a degree equal to the thickness of its protein molecule, and that the finest threads represent a single chain of molecules. THE SILK GLANDS The silk glands of spiders are secreting organs located within the abdomen. Differing in size, form, and location, these organs are classified largely on the basis of their physical characters. Thus, the pyriform glands are pear-shaped, the aciniform are berry-shaped, and the other kinds are similarly identified by their contour. At least seven distinct kinds of glands are known to occur in the whole group of spiders, but not all of them are found in any single family. SILK SPINNING AND HANDIWORK 57 The cribellar glands are found only in spiders that have a cribellum— a flat spinning plate— and are used in conjunction with the calamis- trum, a comb of hairs on the hind metatarsi. The comb-footed spiders of the family Theridiidae possess all six of the remaining types of glands, and are the only ones having lobed glands, which secrete the material of the swathing film. These spiders thus are provided with one more set of glands than their close relatives, the sedentary orb weavers and the linyphiid spiders. Even the oversimplified classifications of Apstein and others demonstrate conclusively that the spinning organs and glands of spiders are the most complicated structures known for the produc- tion and utilization of silk. The several types of glands and the uses of their silk products are enumerated below: /. The aciniform, or berry-shaped glands. These glands are found in all spiders and are characterized by their nearly spherical shape and resemblance to various berry fruits, such as a raspberry. Four clusters, each containing from a few to as many as a hundred glands, send the silk through each of the posterior and median spinnerets. The swathing band is a product of these glands. Ac- cording to Apstein, they also produce the ground lines for the viscid drops. 2. The pyriform, or pear-shaped glands. Also found in all spiders, these glands occur in two clusters of a few to one hundred or more, and communicate with the front spinnerets. The making of the attachment disks is one of their functions, but they some- times contribute wild threads to the thicker draglines. 3. The ampullate, or bellied glands. Known in all spiders, these usually are present as four large, long, cylindrical glands, but fre- quently there are six, eight, or even twelve. They open through spigots which, when four glands are present, are located on the inner side of each of the front and middle spinnerets. Most of the dry silk of spiders, the dragline being the chief agent, is produced in the ampullate glands. Comstock has suggested that the ground line of elastic silk in the orb weavers is produced by these glands, two of which have been modified for the production of this impor- tant element. The fact that the yellow silk of Nephila is spun from the anterior spinnerets partially confirms this opinion. 4. The cylindrical glands. These long, cylinder-shaped glands are often wanting in males, and are lacking in the Dysderidae and the Salticidae. They number six or more, and open on the inside 58 AMERICAN SPIDERS of each posterior spinneret through a spigot. They produce the silk for the egg sac. j. The aggregate, or tree-form glands. There are six of these irregularly branched, compound glands, opening on the inner sur- face of each posterior spinneret through spigots. From these glands, which are found only in the Argiopidae, Linyphiidae, and Theri- diidae, are produced the viscid drops for the viscid threads of the web. 6. The lobed glands. Found only in the Theridiidae, these are irregular in shape and lobed, opening on the posterior spinnerets through spigots. The swathing film of the family is produced in these glands, which are developed largely at the expense of the aciniform glands. 7. The cribellum glands. These numerous, spherical glands open on the cribellum through many tiny pores. They occur only in the cribellate spiders, and secrete the woof of the hackled band. As is to be expected, those spiders that use many types of silk have the greatest number and volume of glands. The abdomen of the sedentary orb weavers is largely filled by glands; whereas the vagrants are less bountifully supplied. In some males the cylindrical glands are missing, and in many males the other glands are less well developed than in females. Inasmuch as the male's need for some types of silk virtually ceases when he becomes adult, the lack of specific glands is of no great importance. The spider has at its command these various types of silk glands and can call upon them for its many needs. Flexible fingers are the spinnerets: they can be extended, withdrawn, compressed, and manipulated like human hands. The filaments produced are some- times simple threads in multiples of two, but frequently they are composite lines and are drawn from different glands. The viscid spiral of the orb-weaver snare, for example, is composed of a double ground line, possibly coming from the aciniform glands, on which is superimposed a thin coating of viscid silk from the aggregate glands. Only when this line is spun in a particular way does it take on the characteristic form of a beaded necklace. The spiral is spun rather slowly, and the spider pulls out the coated line and lets it go with a jerk. As a result, the fluid is arranged in globules, spaced along the line and far more sticky than a thin, uniform covering. The rate of pull and the degree of the tension determine the finished product. The spider spins leisurely or swiftly, according to its need. SILK SPINNING AND HANDIWORK 59 THE DRAGLINE No better illustration of the dependence of spiders on silk is afFored than the habit of laying down a dragline or securing thread. Wherever the spider goes, it always plays out behind from its spin- nerets a silken line, which is anchored at intervals (by means of the attachment disks) to the substratum, as the climber lets out a rope when he enters the recesses of a deep cave or moves down the slope of a precipitous mountain. The dragline is a constant companion of spiders of all ages and all kinds, excepting a small group of primitive forms of the family Liphistiidae. It is the fundamental thread of most spinning. The sedentary orb weaver, committed largely to an aerial life in the confines of its web, outlines the zones of its snare with this thread. Long strands floated in the air form bridge lines from tree to tree or across streams. On draglines, the spider balloons for long distances. Great sheets and flakes of gossamer are mostly the discarded draglines from many spiders. The orb weaver again, hid- den in its leafy retreat, holds a trap line and uses it to detect the presence of an insect in the web. The dragline is the lifeline of the spider. It is an aid in prevent- ing falls from precipitous surfaces, and may also serve as a means of escaping enemies. Web spiders often drop from their webs on these lines and hide in the vegetation. Or they drop down and hang suspended in midair until the danger is past, whereupon they climb up hand over hand to their original position. The hunting spiders jump headlong over cliffs or leap from the sides of buildings to escape capture, and float down gently on their silken ropes. Most of the spinning in our houses is dragline silk, which the house spiders lay down in great profusion and which soon is transformed into the familiar cobweb, heavy with air debris. Even the framework of the retreats is put up with dragline silk, and on this base other types of silk laid. Not a single filament, as the name implies, the dragline in its simplest form is composed of two relatively large threads that ad- here so closely together that only one line is apparent. On occa- sion, the dragline may be made of four strands, or even of a great many threads drawn from several spinnerets. 60 AMERICAN SPIDERS SPIDER THREAD IN OPTICAL INSTRUMENTS The use of spider silk for reticules in various optical instruments is a direct consequence of the fineness of the fibers and of their great strength and ability to withstand extremes of weather. Prior to World War I, spider silk was very extensively used for cross hairs and sighting marks in a great variety of engineering, labora- tory, and fire-control instruments. For transits, levels, theodolites, astronomical telescopes, and many other optical devices there is nothing much superior to spider silk. Most people who use such instruments are familiar with the fibers, and often replace them in the field, using old spider silk or drawing a supply from living spiders. Since World War I there has been a slackening in the use of this material. The finest threads are useless for cross hairs because of their fragility and the difficulty of installation. Because dragline silk is most often used, the joined fibers must first be separated so that the primary line will be a single uniform thread. This can be easily done, since the two or four threads are discrete, and the resultant single strand, averaging 1/20,000 of an inch in diameter, is usable. Even finer fibers can sometimes be used. But the lines spun by spiderlings and small spiders, as well as the finer fibers of larger ones, are usually quite useless. The cocoon silk of the large Argiopes can often be employed for telescopes. The floss beneath the tough outer covering is pulled out easily, and single strands of considerable length procured. This cocoon silk is spun from different glands and is not quite as strong as the dragline silk, which is the most commonly used fiber. The silks of many spiders are suitable for reticules. In Europe the favorite species are large orb weavers such as Aranea diadema and Zilla atrica. Many other spiders provide suitable silk, even those belonging to quite different families. In the United States most silk comes from the common house orb weavers, Aranea foliata and dumetorum, from the numerous humped araneas, from the argio- pids, particularly Argiope aurantia, and many others. The silk of the black widow has also been used extensively. Silk is usually reeled from the spinnerets of living spiders and placed upon suitable frames for storage. It is easy to secure and retains its properties for many years. During World War II there was an increased demand for spider fiber for laboratory and sur- PLATE VII i Walker Van Riper a. The male after mating is occasionally, as here, killed and eaten by the female Walker Van Riper b. A female in her tangled snare with long-legged spiders, Psilochorus BLACK WIDOWS, Latrodectus mactans P LATE VIII Lee Passmore . A desert solpugid (Eremobates) A giant-tailed whip scorpion, Mastigoproctus giganteus RELATIVES OF SPIDERS SILK SPINNING AND HANDIWORK 61 veying instruments. Although few of the optical instruments re- quiring spider silk were directly concerned with war in the field, some newspaper publicity gave the impression that the silk was in great demand as a critical war material. The truth of the matter is that all needs were satisfied by a few individuals who only devoted part of their time to the securing of the web. The importance of spider silk in industry has decreased progres- sively during the past thirty years. Its place has been taken by platinum filaments and by engraving on glass plates. Where an aerial reticule is desired, drawn filaments of silver-coated platinum wire are frequently used. These filaments, usually 1/10,000 of an inch in diameter, are mounted in a heavy metal ring to form the desired pattern. They are said to be superior to spider web since they show an even black line and do not sag in a humid atmosphere. For all instruments requiring a complicated pattern, etched glass reticules are usually used. In bomb sights, range finders, periscopes, and most gun sights, in fact in virtually all optical fire-control in- struments, the width of the line has to be carefully adapted to the optical purposes and characteristics of the instrument. Etched glass is obviously necessary in most such instances; it would be impossible to accomplish the desired results with spider silk. SILK FOR TEXTILES It has for centuries been the ardent desire of araneologists to find some way of exploiting for commercial purposes the tremendous supply of spider silk available in nature. As long ago as 1709, a Frenchman, Bon de Saint-Hilaire, demonstrated that spider silk was usable for fabrics in the same way as the silk of the silkworm. A large number of egg sacs were washed, boiled, and cleansed of all extraneous matter, then allowed to dry out. With fine combs the sacs were carded and worked into slender thread of a pleasing gray color. Two or three pairs of stockings and gloves were made from the natural silk, and were presented to the French Academy. So sensational was this accomplishment that in 1710 the Academy of Sciences of Paris commissioned R. A. de Reaumur to investigate the possibility of an extensive utilization of spider silk. After a thor- ough study, this eminent entomologist (and inventor) concluded there was little likelihood that spider silk, at least such as was avail- able in Europe, could become a profitable industry. 62 AMERICAN SPIDERS The difficulties he enumerated are inherent in the spiders them- selves and in their silk, and are still those that rule out the silk of spiders as a potential material for commerce. In the first place, spiders are solitary, predaceous animals that feed only on living invertebrates. Each spider must be segregated and maintained apart from its neighbor, for cannibalism is the rule when the larger spiders come together, and the population is soon decimated. Space requirements are considerable; the difficulties of providing suitable food are almost insurmountable. Only the egg-sac silk was con- sidered at that time to be usable, and, although many sacs are pro- duced by the females, it would require, as de Reaumur estimated, 663,522 spiders to produce a pound of silk. On such terms, compe- tition with the silkworm was impossible. The silk itself was considered inferior in strength to that of the silkworm, owing to its far finer threads, which lacked the luster of insect silk and were difficult to work satisfactorily. The silk- worm produces a single line of silk, which is usually between four and seven hundred yards long— a production representing the total output, the whole lifework, of the larval moth. Even with the relatively thick lines of the silkworm, their joining together to form commercially usable threads is an exacting process, which, because no mechanical solutions have been successful, must be done by hand. Strands of spider silk do vary in thickness, and the large silk spiders of the genus Nephila, which abound in the East Indies and in the Orient, produce a silk noted for its strength. However, state- ments that the lines in the webs of Nephila sometimes attain the thickness of darning wool are exaggerations. Their thickest line is very much finer than that of the silkworm. In Madagascar an attempt was made to take silk from the local spiders by drawing it directly from their bodies. The natives brought the animals into cleared areas and established them in great numbers near the site of the reeling apparatus. At intervals, the mature spiders were removed from their webs and imprisoned in a most curious device consisting of little stocks that held them firmly between cephalothorax and abdomen. Then small revolving mills were touched to each spinneret, and, as the filaments were pulled out, they were rolled into a single thread by a hand-operated mill. The silk so produced was of a beautiful golden color and quite as good as that of the silkworm, but the project had to be aban- doned because of the practical difficulties. In the United States, Dr. B. G. Wilder drew attention to the SILK SPINNING AND HANDIWORK 63 possibility of using the silk of the big American Nephila. In 1866 he extracted silk directly from the body of this spider— unaware, at the time, of the earlier European experiments. Wilder was amazed by the ease with which it was possible to reel off the golden silk, and intrigued by the possibility of producing quantities of it for textiles. From one spider he reeled off silk for an hour and a quarter, at the rate of six feet per minute, taking a total of one hundred and fifty yards. Later he devised an ingenious little apparatus to hold the spider during the reeling, and was able to obtain quickly the full quota of available silk. In addition to holding the creature firmly in stocks, the device had a round piece of cork on which the spider could rest its legs, thus being prevented from interfering with the flow of silk from its spinnerets. Dr. Wilder found that one female would yield at successive reel- ings one grain of silk, and that four hundred and fifteen spiders would be required to yield one square yard of commercial silk. For an ordinary dress requiring twelve yards of material, therefore, nearly five thousand spiders would be required. This was quite bountiful production for spiders, yet it is still only half the amount obtainable from an equal number of silkworms. Today we are no nearer than Saint-Hilaire and Wilder to a realization of spider silk as a practical commercial textile. The basic obstacles remain, inherent in the characteristic differences be- tween the silk spider and the silkworm. USE OF SILK BY PRIMITIVE PEOPLES A material of such abundance and strength as spider silk could scarcely have failed to be used by primitive peoples for some of their needs. Indeed, it is surprising that we do not have more records of its use in the Americas, where the same types of spiders abound that have supplied the Papuan and Oriental natives for gen- erations. From the great Nephila spiders comes silk to supply cer- tain New Guinea natives with gill nets, kite nets, dip nets, and va- rious lures for their fishing activities, silk with which to weave bags, caps, and headdresses, and silk for other purposes. Strength resides not in a single strand of silk but rather in the twisted and matted threads, which form a tough fabric. The large aerial webs of Ne- phila are made with a very strong silk, and are capable on occasion of ensnaring birds in their viscid and elastic lines. 64 AMERICAN SPIDERS In the New Hebrides, the natives use spider silk to fabricate small bags in which they carry arrowheads, tobacco, and even the dried poison used on their arrowheads. Some New Guinea natives of the Aroa River district make a headdress of insect or spider silk to keep out the rain. To more sinister uses are put the smoth- ering cap and the dooming bag, both made by the New Hebrideans. The former is a strong, conical cap which is pulled down tightly over the heads of victims, usually adultresses, and causes death by suffocation. The dooming bag, a purse filled with various bric-a- brac, is said to have magical properties. According to the stories, it is rubbed over the forehead of a sleeping victim with a rhythmic motion and with muttered magical words, causing him to remain in a deep hypnotic sleep from which there is no awakening. The soporific effect of the dooming bag is assured by the victim's exe- cutioners, who administer a coup de grace after they have carried him into the jungle. Of more interest are the fishing nets of the Papuans, which show varied and ingenious use of spider fiber. Several accounts illus- trating primitive man's ability to seize upon common materials and suit them to his purposes are well worth mentioning. The North Queensland black boy entangles one end of a thin switch in the web of Nephila and, by adroit weaving motions, twists the coarse lines into a strand a foot or more long. The frayed ends of the line are moistened in the crushed body of the large olive-green silk spider (known to these aborigines as "karan-jam- ara") and the remaining morsels are thrown into the stream, imme- diately attracting shoals of small fishes. As the silken lure is trailed through the shallow water, a fish rises to sample the tidbits on the invisible strand. Lines of gossamer become entangled in its teeth, and the smiling angler lands the two-inch long prize with a careless flourish. This method of fly fishing, and other engaging fishing techniques of the Australian aborigines, may be found described in detail in E. J. Banfield's book Tropic Days. That the catch is limited to small fishes does not detract from the efficiency of the method. Many are caught in a relatively short time, seventeen fingerlings in ten minutes according to one account, and make up in numbers what they lack in size. It is said that these lures, as generally made, are capable of holding fish weighing nearly three-quarters of a pound. A similar lure is used as part of a novel method of catching fish on the east coast and adjacent islands of New Guinea, and in the PLATE 9 Richard L. Cassell Crab spider, Misumenoides aleatorius, on flower PLATE IO a. Huntsman spider, Heteropoda venaloria J. M. Hollister b. Silk spider, Nephila clavipes SOUTHERN SPIDERS SILK SPINNING AND HANDIWORK 65 Solomon Islands. The natives make a kite of the large flat leaves of one of the local trees, sewing them together and stiffening them with tough strips to produce an object about two and a half feet long and nearly a foot in width. The completed kite is embellished with five wings of pandanus leaf. A flying line is made of fiber twine, ordinarily about one-third of a mile long, while the tail is another length of twine from one to three hundred yards in length, at the end of which is tied a tassel made from the web of silk spiders. The kites are then flown over the seas either from the shore or from canoes in such a way that the spider tassel skips along the water and entices fish to strike. The golden-yellow silk entangles the teeth of the fish, and, after some maneuvering with kite and boat, it is lifted into the canoe by means of a dip net. Still another intriguing method of capturing small fish is prac- ticed by certain Solomon Island natives. This account by H. B. Guppy is from The Solomon Islands and Their Natives: The following ingenious snare was employed on one occasion by my natives in Treasury, when I was anxious to obtain for Dr. Gunther some small fish that frequented the streams on the north side of the island. I was very desirous to have some of these fish, and my natives were equally anxious to display their ingenuity in catching them. They first bent a pliant switch into an oval hoop about a foot in length, over which they spread a covering of stout spider-web which was found in a wood hard by. Having placed the hoop on the surface of the water, buoy- ing it up with two light sticks, they shook over it a portion of a nest of ants, which formed a large kind of tumour on the trunk of a neighboring tree, thus covering the web with a num- ber of struggling young insects. This snare was allowed to float down the stream, when the little fish, which were between two and three inches long, commenced jumping up at the white bodies of the ants from underneath the hoop, apparently not seeing the intervening web on which they lay, as it appeared nearly transparent in the water. In a short time, one of the small fish succeeded in getting its snout and gills entangled in the web, when a native at once waded in, and placing his hand under the entangled fish, secured the prize. With two or three of these web hoops we caught nine or ten of these little fish in a quarter of an hour.3 8H. B. Guppy, The Solomon Islands and Their Natives, London, 1887, p. 157. 66 AMERICAN SPIDERS The Papuan natives make landing nets from the orb webs of Nephila. A. E. Pratt describes this practice as follows: One of the curiosities of Waley (near Yule Bay), and, in- deed, one of the greatest curiosities that I noted during my stay in New Guinea, was the spiders' web fishing-net. In the forest at this point huge spiders' webs, 6 feet in diam- eter, abounded. These are woven in a large mesh, varying from i inch square at the outside of the web to about Y8 th inch at the centre. The web was most substantial, and had great resisting power, a fact of which the natives were not slow to avail them- selves, for they have pressed into the service of man this spider, which is about the size of a small hazel-nut, with hairy, dark- brown legs, spreading to about 2 inches. This diligent creature they have beguiled into weaving their fishing-nets. At the place where the webs are thickest they set up long bamboos, bent over into a loop at the end. In a very short time the spider weaves a web on this most convenient frame, and the Papuan has his fishing-net ready to his hand. He goes down to the stream and uses it with great dexterity to catch fish of about i Ib. weight, neither the water nor the fish sufficing to break the mesh. The usual practice is to stand on a rock in a backwater, where there is an eddy. There they watch for a fish and then dexterously dip it up and throw it on the land. Several men would set up bamboos so as to have the nets ready all together, and would then arrange little fishing parties. It seems to me that the web resisted water as readily as a duck's back.4 Although Pratt's account has not been verified, there is never- theless more reason to believe that it could be true of Nephila webs rather than of the garden variety of orb web. It is not difficult to persuade the spider to use a bamboo hoop, since it is a most suitable framework for a web, and we know that American orb weavers sometimes oblige by spinning a web on a frame supplied them. It is also true that the radii of the Nephila webs are more numerous, and that the many closely set spirals would contribute to the strength of the web. The spiral line becomes a permanent part of the web and thus multiplies its strength. Finally, it is possible that immersion in water contributes in a mechanical way to the strength, making the struggles of the fish less liable to rupture the lines. *E. A. Pratt, Two Years Among New Guinea Cannibals, London, 1906, p. 268. SILK SPINNING AND HANDIWORK 67 In other reports of spider-silk landing nets, the spiders are not said to spin their webs upon the hoops, but instead the latter are twisted and turned through a number of large webs until there results a many-layered covering of fiber. When drawn through the water, these nets are stretched to the shape of a shallow bag. In the Trobriand Islands, the frame is only the fork of a shrub on which the web is twisted. Apparently, these homemade landing nets are quite durable objects and can be used over and over again, ordinarily requiring only a limited amount of patching. Their haul is made up of prawn, shrimps, and various kinds of fish sometimes weighing as much as three or four pounds. Thus the wiles of the modern angler, with his casting fly, his trolling lines, and his gill nets, are duplicated by natives of a single region, and the important element of all these snares is spider silk. CHAPTER V Courtship and Mating OPIDERS HAVE GONE TO GREAT lengths to ensure that the chain of life continues strong and un- broken. The bringing-together of the sex cells is accomplished by these arachnids in a manner so extraordinary that the various strange details almost transcend belief. Completely lacking a primary in- tromittent organ at the site of the genital opening, the male spider has developed a secondary one of wonderful complexity at the end of each of the pedipalpi. The female has also developed, in comple- ment to the palpi of the male, an organ called the epigynum, which lies immediately in front of the genital opening and which is spe- cialized to receive the male palpus, to store the sperms, and to com- municate them to the liquid body of the egg mass 'at the time of egg laying. The roles of the male and female preparatory to and during the mating are quite distinctive. Soon after the male become mature, following the last molt, when the palpal organ is completely de- veloped, he transfers to the palpi the sperm produced in the testes. This step is termed sperm induction. Next he must find a female and overpower her, or must elicit through characteristic actions an acquiescence to his desire for mating. The series of actions that mark the period during which he is endeavoring to gain her recog- nition is called courtship. The mating itself is accomplished by means of a series of accessory apophyses on the palpi, on the legs, or on other parts of the body, which seize and orient the bodies of both sexes in such a way that the palpal organ can come in contact with the apparatus of the epigynum. On the other hand, the role of the female is a more passive one; she needs only to ascertain through instinctive means that the presence of the male is to be welcomed and to conform to the distinct pattern that makes possible a successful mating. 68 COURTSHIP AND MATING 69 SEXUAL DIMORPHISM Spiders are to all intents sexless until they arrive at maturity. There is nothing in the general appearance of immature specimens that indicates with certainty femaleness or maleness, and nothing in their early-life activities of digging, hunting, or web spinning that marks either sex. Many people think of immature spiders as being female, and there is good reason for this since they usually more closely approximate the mature female in general appearance. It is probable that ancestral spiders exhibited little sexual dimorphism, and we note that this is true for some (but by no means all) of the more primitive types. Changes in the sexes have occurred both in the female and male, but they have been far greater in the male. The female is specialized for a particular function, and, if we presume to evaluate the sexes in finite terms, is a far greater con- tributor to the race than the male. Whereas the male has completed his assignment when he transfers the sperms to the female recepta- cles, the female maintains the eggs in her body until they are ready to be delivered and fertilized, encases them in a silken sac, guards them in various ways, and often is on hand to protect the young spiderlings for a considerable period. Her body has been molded as a receptacle for nurturing a variable number of developing eggs, and it responds to this need by maintaining a greater size than the male. Perhaps in response to her protective role, she is less con- spicuously colored and far less of an extrovert than her male. On the other hand, because of greater size, she is much more powerful; and she is dominated most of the time by predatory instincts inten- sified by her solitary habits. Among spiders, the male is a luxury item, developed for the single purpose of transferring the sperm. He offers no protection to the female or the offspring, as do many other animals, and is usually dead before the eggs are laid. He has changed in various ways to become a specialist, and is modified in many ways to play his part more expertly. The force that sends him into the arms of the female ogre is a very strong one, but he has become conditioned to preserve himself by taking flight should he be unwelcome. He has also become conditioned to overpower the female on certain occasions. The specialization of the male has proceeded in several direc- tions, and we find a considerable variety of types. In many of the 70 AMERICAN SPIDERS hunting spiders the sexes are quite similar in size and seemingly nearly equal in strength. But even with these there are noticeable differences. The abdomen of the male is slimmer, and frequently clothed with somewhat different hairs and patches of setae. The color pattern of the males is almost always somewhat brighter, even though the species be classified as drab. In these spiders of nearly equal size (the Lycosidae, Oxyopidae, Gnaphosidae, Clubionidae, and others), the outstanding feature of the male is his somewhat longer legs, which give him a greater range of sensory perception and are thus important in evading and overpowering the female. This disparity in leg length is presumably maintained because of and correlative to the quite different modes of life of the sexes, and the dedication of the whole adult life of the male to sex. Among the spiders that have quite similar sexes except for the longer legs of the males are the trap-door spiders. We can interpret in various ways the difference in leg length. In addition to the advantages enjoyed during the courtship and mating, it may mean that the male is better fitted to wander about in search of the female. On the other hand, the longer legs may represent the more generalized condition, and the shortening of the legs of the female a response to the burrowing habit. Sexual dimorphism manifests itself in pronounced difference in size in many of the higher web spiders. Among the orb weavers exist all intergrades between a near equality of the sexes and a reduc- tion of the male size to an infinitesimal portion of the female bulk. The large humped orb weavers have males that are nearly equal to their females, but in other members of the same genus Aranea the male may be one fourth her size. In Argiope (Plate XVIII), Cy- closa, and many other genera, the male is much smaller than the female, in the first genus being about one fourth as long. The dis- parity is far greater in Nephila, where the female of the American species weighs more than one hundred times as much as the male, and in some exotic species is said to be over one thousand times larger than the male. The male is also a pygmy among such spiders as Mastophora (Plate III), Gasteracantha, and Micrathena. A re- markable sexual dimorphism exists also among the comb-footed aerial spiders, the Theridiidae, and the vagrant crab spiders of the family Thomisidae. The smaller size of the male gives it certain advantages during courtship and mating, and perhaps is used to counterbalance the physical superiority of the female. In Mastophora and Nephila it COURTSHIP AND MATING 71 has been carried to a ridiculous extreme. These tiny males are vir- tually immune to the attacks of the great females, being far beneath the usual size of the latter's prey. Tiny insects have much the same immunity, and are tolerated when they crawl over a spider's body and left untouched when they are caught in its web. Great reduc- tion in size doubtless represents an orthogenetic development that has nothing to do with the needs of the sex itself, but persists once it has started. It also brings with it other problems, since the males become sexually mature weeks in advance of the females and must live until the females mature. The males possess the two pedipalps with the complex sexual organs at the end; these organs have become wonderfully specialized to aid in the pairing. The legs are also frequently armed with spurs or with rows of modified and enlarged spines that aid in clasping the female or in holding her chelicerae or appendages. The taran- tulas, trap-door spiders, and many of the primitive true spiders have two prominent processes on the front legs to catch the appendages of the female. The elongated chelicerae of Tetragnatha and Pachy- gnatha are used to grasp those of the female. Among aerial sheet weavers, the Linyphiidae, we find a tremendous group of species in which the heads of the males are specialized in divers peculiar ways. There are pointed or rounded spurs armed with curious setae, great rounded lobes, long, thin processes, prolongations of the clypeus or front; and the eyes are often carried to the tops or sides of these eminences. There is little doubt that these spurs are of significance in the mating of the species. In some of these spiders, it is known that the female fixes her chelicerae in the tiny pits on each side of the head lobe, and thus orients the male for mating. Sexual dimorphism also manifests itself in profound differences in color pattern and intensity. The carmine legs and shining black body of At y pus bicolor, the large purse-web spider, far outshine the pleasant brown tones of the female. The male tarantulas have a darker body and often have the abdomen set with long golden or reddish hairs. Among the true spiders, the males are much more varied and usually more handsome than their mates. This is espe- cially true of the jumping spiders, which in the tropics display a spectrum of color, the most brilliant hues of which are restricted to the males. In many instances, the sexes are so different in appearance that they were formerly regarded as being of distant species. Among certain of the sedentary spiders the sexes are somewhat more equal from the color standpoint; and of the spiny-bodied 72 AMERICAN SPIDERS spiders, Gasteracantha and Micrathena, the females even have beau- tifully painted and sculptured bodies. SPERM INDUCTION The strange process by which the male spider transfers semen from the primary genital organs into the receptacles of the palpi is called "sperm induction." It was observed for the first time in 1843, by Anton Menge, who described how the male constructed a little web of silk, desposited a droplet of sperm upon it, and then applied his palpi to the drop until it was entirely absorbed into these latter organs. It is not at all surprising that this extraordinary action was doubted at first by many people, among them several eminent arachnologists, who insisted that there must be an internal connec- tion between the testes, deep in the abdomen, and the tips of the palpi. Now we recognize sperm induction as only the first step in a series of strange acts that mark the sexual life of spiders. Sperm induction is of necessity a very common phenomenon, but one must be on hand at the right time to observe it. Soon after the male becomes sexually mature, he charges his palpi and is then ready to wander about in search of a mate. This is an act which is not part of the previous experience of the male, but' is initiated by internal changes in the body associated with the arrival of maturity. He performs it instinctively and perfectly at the outset, because it is fixed in his behavior as a racial memory. Thereafter, he fills his palpi frequently, usually immediately following copulation, which is the best time to see this interesting spectacle. A few spiders are able to mate more than once without exhausting their semen; others have to pause during their mating to refill the bulbs. There is considerable diversity in the manner in which different types of spiders accomplish sperm induction. In no known instance is the sperm taken directly from the genital opening at the base of the abdomen, which would appear to be a logical means of solving the problem, and would be physically possible in many spiders with long palpi. Some spiders spin very simple, loose webs and absorb the semen by placing the palpi directly against it. In Pholcus a single silk line between the third legs is drawn across the genital opening until the spermatic globule adheres to it, whereupon it is taken up by the chelicerae and held there for direct absorption by the palpi. Some of the other primitive spiders do essentially the COURTSHIP AND MATING 73 same, but hold the globule and the tiny web between the palpi or front legs until it is absorbed. A great many spiders spin a tiny sheet of very fine web, usually quadrangular or triangular in outline, place a drop upon the surface, and then take the semen indirectly by applying their palpi on the opposite side of the sheet. Among the tarantulas sperm induction is a long operation that sometimes requires three or four hours. The male spins a large flat sheet of silk, attached to adjacent objects, in which are left a large oval opening and a much smaller one, the two separated by a nar- row band of strongly woven silk. He then crawls through the large opening, and, lying on his back, strengthens the silk around the holes. After rubbing his palpi through his chelicerae and stroking his genital opening against the reinforced silk band between the holes, a drop of spermatic fluid appears and is deposited on the underside. The male now clambers back, and, sitting upright on the web over the globule, reaches around the edge of the narrow band to touch the sperm directly. The process of absorption takes an hour or more, and consists of a rhythmical alternate tapping of the palpi in the globule, usually at the fast rate of one hundred to one hundred fifty per minute. Afterward the web is destroyed or de- serted. Most spiders are able to recharge their palpi much more quickly, usually within half an hour. T. H. Montgomery has described how one of the small American wolf spiders, Schizocosa crassipes, spun a triangular sheet attached to the floor and walls of its cage, and stood on the upper side of this web. A small globule of yellowish semen was ejaculated upon the surface of the sheet at about the middle. The male "then reached his palpi downward and backward, below the sheet, and applied the concave portion of the palpal organ of each against that part of the sheet which carried the drop of sperm. Each palpus was then rubbed against the lower surface of this drop several times, then withdrawn and slowly shaken in the air, while the other was similarly applied to the drop." 5 This process con- tinued for seven minutes, during which time all the sperm was taken up into the palpal organ, and soon afterward the male left the sperm web. The male seems to derive considerable gratification from the process of sperm induction. Before the act, the genital region is rubbed against the strands of webbing to incite the ejaculation of $T. H. Montgomery, "Studies on the Habits of Spiders, Particularly Those of the Mating Period," Proc. Acad. Nat. Set., Philadelphia, 1903, p. 65. 74 AMERICAN SPIDERS the semen. The presence of the female is not a necessary adjunct of the act, which is implemented by internal factors, whereas later on she, or her threads, become the stimuli which result in the mating. COURTSHIP Inasmuch as the young male leads the same kind of life as the fe- male, and lives in similar webs on plants or hides in similar places on the ground, maturity finds him not far distant from female neigh- bors. After he has prepared himself for mating by charging his palpi, a new impulse sends him in search of a female of his species, and he moves about in a random manner until he is able to detect his mate. Since spiders are largely creatures of touch, it is not surprising that to find the female he relies mainly on the fine sensory hairs that clothe his body and appendages. Contact with the substratum brings him something more than the mere mechanical sensation of touch or tension or vibration. Accompanying it is an ability to dis- tinguish certain chemical substances with which his hairs come in contact; this combined sense is called chemotactic. The receptors for it have not been recognized, but it seems reasonably certain that some of the hairs on the appendages are sensillae that respond to this type of stimulation. Since the sensation comes to the spider only when in actual contact with chemical substances, it is nearer that of "taste" than "smell," but it remains a quite different sense from any possessed by man. The male spider thus becames keenly aware of the presence of a mate through the touch of her threads, or of the trail she leaves on the substratum, or of her actual body. There is still another way in which some spiders are able to dis- cover their mates. The vagrant spiders have developed eyes of such acuity that they can see moving objects at a considerable distance— for a spider— and can identify the other sex when still several inches away. In these relatively long-sighted types, and especially in the jumping spiders, recognition of the female may be possible by sight alone, without any aid from the chemotactic sense. On the other hand, certain of the wolf spiders, having vision nearly on a par with the jumpers, nevertheless appear to require both sight and touch to incite pairing. Once the male has dicovered the female, he is on the threshold of realizing the racial instinct for which he has become specialized— COURTSHIP AND MATING 75 the transfer of the semen. But there are difficulties. The object of his attention may not be of the same mind as he is, and she usually exceeds him in size and strength. Further, since virtually all of her life has been devoted to capturing and feeding on animals of suit- able size, her first instinct is predatory. That the interloper is a male of her own kind is immaterial, if she is not conditioned to distin- guish him from any other suitable prey. There consequently ensue certain more or less marked preliminary activities before the actual mating, which constitute the spiders' courtship. Most of the initia- tive is taken by the male, who— being the less valuable sex— is con- ditioned to make the first advances and brave the danger. Upon him rests the burden of announcing himself in a convincing manner, and of stimulating the female to a point where sexual union is possible. Among the aerial spiders and other web spinners, courtship usu- ally consists first of some kind of vibration of the threads of the web, and later of stroking the body of the female. Among the hunting spiders there is a considerable diversity in methods of courting. The species blessed with good eyesight have developed a relatively complicated prenuptial procedure in the course of which the male advertises his presence by movements of the legs and body. Correlated with this behavior to some degree are various epigamic structures such as brushes or ornaments on the legs and tufts of hair on the head. Spiders with poorer eyesight are ordi- narily much more conservative in their prenuptial routine, since the female would be unable to see the details; but occasionally body ornaments are present in this group as well. There are numerous intergrades between a well-marked courtship, as exemplified in the bizarre love dances of the jumping spiders, and almost no courtship at all; the fundamental mechanism and the particular path that each group has followed to arrive at its present specialization are subjects that must be discussed at length. The prime descriptive and analytical studies of spider courtship and sexual biology, following the classical work of Anton Menge, were made in the United States by G. W. and E. G. Peckham in 1889 and 1890, and by T. H. Montgomery in 1903 and 1910. In addition to fascinating descriptions of the sexual processes in many species, quite adequate explanations of the significance and evolution of various phenomena in terms of selection were presented. The Peckhams were strong exponents of sexual selection as outlined by Darwin, and concluded that the female jumping spider responded to the charm and beauty of the posturing male and made a conscious 76 AMERICAN SPIDERS selection of a mate. They believed that the males were more numerous and, especially in cases where there was male dimor- phism, that the brighter male was preferred by the female. They argued that the numerous ornamental features on the bodies of the male jumping spiders were developed as a result of sexual selection. They rejected A. R. Wallace's views that such epigamic characters were a result of a surplus of vital energy that went with maleness: because the male was more vigorous, he was more highly colored and likely to be more successful in his suit with the female, and thus would more surely and more often leave progeny. In 1910 Montgomery rose to the defense of ordinary natural selection, and in a masterful essay virtually refuted the claims of the Peckhams with regard to sexual selection. Montgomery be- lieved that the adult male "is excited simultaneously by fear of and desire for the female, and his courting motions are for the most part exaggerations of ordinary motions of fear and timidity. By such motions he advertises himself to the female as a male, but there is no proof that he consciously seeks to arouse her eagerness by esthetic display . . . there seems to be no good reason to hold that the female is actuated in her choice by sensations of beauty." 6 Montgomery defined courtship in a more limited way than do modern arachnologists, and believed that in some vagrants there was none at all. However, judging from his descriptions, his interpre- tation is in most cases a modern one. Commenting upon spiders that have good sight, he said as follows: "What we do know is that the male by his courtship, a set of motions resulting from the conflicting states of sexual desire and fear, exhibits or advertises himself as a male; and that the female on sight of this courtship recognizes him as a male and accepts him if she be eager, or else becomes gradually stimulated by watching him." 7 Montgomery further believed that many secondary sexual characters in the male "may be most readily explained as being conserved by simple selec- tion. Peculiar male ornamentation would be selected because it insured quicker sex-recognition, therefore prompter mating. The male is thereby more surely accepted by the female, not selected by her in the sense of Darwin. The process is much more an announce- ment of sex by the male than a choice by the female, and results in ST. H. Montgomery, "The Significance of the Courtship and Secondary Sexual Characters of Araneads," The American Naturalist, Vol. XLIV (1910), pp. 151-2. 'Ibid., p. 169. COURTSHIP AND MATING 77 the female accepting the sex rather than the individual." 8 Mont- gomery did not subscribe to Wallace's belief that the males ex- hibited a higher degree of vitality, but argued instead that the need of greater protection by the females was the reason for their less conspicuous coloration, as in birds. It has remained for W. S. Bristowe to take up the problem where Montgomery left off, and to extend and elaborate his thesis on the basis of a much vaster literature and innumerable observations of European spiders. Bristowe's views were presented in convincing fashion in 1929, in a long paper entitled "The Mating Habits of Spiders, with special reference to the Problems surrounding Sex Dimorphism." In this treatise he pointed out that the complicated visual displays of the jumping spiders probably arose by ordinary natural selection. Primitive spiders were short-sighted hunters that groped their way as they walked and stretched out their front legs to test the substratum. Perception of the environment was accomplished by a chemotactic sense largely confined to the extremities of the append- ages. Since sight was limited, it was necessary for the male to touch the spoor, the threads, or the body of the female to discover her presence. Since the males were able to detect the presence of a mate often before she was touched, those males that started to ad- vertise their identity early by means of their front legs were more likely to survive the assault of their larger, predaceous mates. Aiovement of the appendages and parts of the body enhanced the chances of survival and also increased the possibility of finding a mate. All these advances in posturing were accompanied by a gradual improvement in the acuity of the eyes, likewise arrived at by selection. Males tend to produce more pigment than females, so those that were able to develop strikingly colored spots in front, visible to the female, were able to survive more often. The various antics and decorations worked hand in hand. The most generalized types of courtship are exhibited by those spiders in which distance perception is feebly developed, the ma- jority being short-sighted hunters. More specialized displays have arisen in two ways: By improvement in the acuity of the eyes, as in the long-sighted hunters; and by development of expansive webs that enlarge the limits of perception by touch, as in the web build- ers. These divisions approximate in a general way the systematic position of the species. ., p. 173. •Ibid. 78 AMERICAN SPIDERS The Short-Sighted Hunters. Most of these spiders are nocturnal creatures of the ground that rely almost entirely on touch to inform them of the character of their environment. They test the surface by means of their legs, which act both as organs of touch and of- fensive weapons. Their approach to the female is usually a bold one, since most of them approximate her in size, and a mere touch is sufficient to inform them of her sex and species. Further, this physical contact probably gives the female the same information. Recognition is almost instantaneous and largely based on the chem- ical sense. There remains only for the male to stimulate the female until mating can be accomplished. He does this by stroking and tickling her body, while at the same time maintaining a firm grip on her with his legs or chelicerae, so that she cannot escape. The tarantulas are wonderful subjects for the study of mating behavior in the short-sighted spiders. Alexander Petrunkevitch has described the courtship of Dugesiella hentzi thus: When the restlessly wandering male happens to touch with his legs some part of the body or a leg of the female, he at once stops short and begins to strike simultaneously and violently with his anterior, sometimes with all four front feet. . . . This continuous beating with the front legs upon the body or legs of the female constitutes the first step in the courtship on the part of the male. In case the female does not attempt to run away, the male soon shifts his position until he is facing the female. The behavior of the female during the first stage of the courtship is composed of two elements. At the first touch she raises the front legs and assumes the attitude of defense and threat. The subsequent touching results in her rising high on her hind legs while still holding up the front legs. Finally, she opens the fangs and the male catches them with the hooks on his front legs. . . . They serve admirably to guard the male against possi- ble injury or even death while at the same time aiding him in the act of coitus. For he now forcibly pushes back the cephalo- thorax of the female with his front legs and drums with the patellas of the palpi on her sternum, all the time advancing.9 The mating that follows lasts only a minute or two, after which •A. Petrunkevitch, "Sense of Sight, Courtship and Mating in Dugesiella hentzi (Girard), a Theraphosid Spider from Texas," Zool. Jahrbucher Syst., Vol 31, p. 373. COURTSHIP AND MATING 79 the two sexes part, the female ordinarily making no attempt to attack the male. Many of the small running spiders spin little silken cells under stones or in tiny nooks on trees. Wulfila saltabunda, one of the smaller anyphaenids, weaves a little curtain beneath the leaf of an herb or bush and stands upright on the silk. The male stands be- neath the sheet and drums on it with his long front legs and palpi, and at intervals pulsates his abdomen up and down. The female often responds by tapping with her front legs and palpi, and vi- brates the sheet immediately above the male. The male will court the female in this position for hours, but mating ordinarily does not occur until evening. He seems able to avoid the female with- out great effort, and to be relatively immune to her attacks. How- ever, she is much more powerful and can kill him with ease if he approaches her too insistently when she is pregnant or otherwise not ready for mating. Some of the gnaphosids, notably Drassodes and Zelotes, are said to take possession of an immature female by enclosing her in a silken cell. Just after her final molt and before she has attained her full strength, the male mates with her. This is possible since the male matures earlier than the female and is able to recognize her as a prospective mate even though she is immature. It is the habit of many of these spiders to live in adjacent silken sacs under the same stone or piece of bark. Not uncommonly, a male in the penultimate stage, when he presumably has no instinct for recog- nizing or sequestering a mate, will be found in a sac with an im- mature female. This suggests that the association in many instances may be only a fortuitous one. A few of the sedentary spiders with inferior eyesight have given up life on webs and have become vagrant secondarily. The most interesting example is that of Pachygnatha, one of the big-jawed spiders that live in grass and vegetation especially in cattail marshes. The male prowls among the grass roots and finds his mate by touch. He seizes her and, aided by special spines and long teeth on his chelicerae, holds hers firmly until her mating instincts have become aroused or her hostility forces a retreat. Among the crab spiders we find few of the preliminary activi- ties identifiable as true courtship. These spiders live on the ground and on plants and are for the most part diurnal in habit. The eyes of some are fairly large, but the spiders seem to make little use of 8o AMERICAN SPIDERS sight in their hunting or courting, a fact which may be partially accounted for by their habits of deception and inactivity. When a male discovers the female of his species, he immediately climbs upon her back or seizes an appendage with his chelicerae. He is much more agile, and can tickle and caress her body until he is able to accomplish his purpose. In some of these spiders the males are very much smaller, and usually more darkly marked than the females. The latter often walk around with a tiny long-legged male clinging to their backs, paying little attention to his activities. Certain spiders have interpolated in the sequence of their court- ship an habitual act that tends to set them apart from all others. The male of the stocky little species of Xysticus spins a thin web over the female, attaching tiny silken lines from her abdomen and legs to the substratum. This web has been called the bridal veil, and its spinning is one of the extraordinary prenuptial habits of many crab spiders. The Long-Sighted Hunters. It is among the spiders of this group that we find those notorious for their love dances (Text Fig. 2). Almost all are day hunters, a habit in keeping with their need for light to display themselves properly during courtship. Some of them are well-known vagrants, and have received such expressive names as "wolf spider," "lynx spider," and "jumping spider" in rec- ognition of their life of action. However, even in this group with the best eyes, reliance is only partially placed on sight during court- ship; and in most instances the event does not occur unless the male actually touches the female, even though he may perceive her by sight. The whole makeup of the prenuptial display— posture, antics, and epigamic ornaments— is distinct for each species. While de- veloping these features nature has had to keep many allied species separate, and thus has evolved by selection many different kinds of dances. The female has become conditioned to respond only to those performed by her species, and rarely makes mistakes. The actual mating is usually preceded by a certain amount of stimula- tion by the legs of the male, and it is this final action that com- pletely precludes the possibility of any related species being ac- cepted. In most wolf spiders, the palpi and front legs are provided with some kind of ornamentation that contrasts sharply with the rest of the body. Where such epigamic characters are present, their PLATE I Richard L. Cassell Black widow, Latrodectus mactans, with prey PLATE 12 Lee Passmore a. Tarantula, Aphonopelma, and tarantula hawk Lee Passmore b. Tarantula, Aphonopelma, and tarantula hawk TARANTULAS COURTSHIP AND MATING 81 movement is usually part of the courtship ritual. Among the small Pardosae the palpi, usually bedecked with jet-black hairs or varie- gated with black and snow-white ones, receive a large share of the ornamentation. The larger lycosids usually have the front legs darkened and occasionally provided with brushes of conspicuous black hairs. The courtship patterns of American wolf spiders were first in- vestigated in detail by Montgomery in 1910, and more recently, in 1936, were made the subject of special analysis by B. J. Kaston. A few of our species may be mentioned. Pardosa milvina (Plate XXV and Text Fig. 2, A) is a small, long-legged wolf spider with black head and palpi. The courtship motions are as follows: The male stands with his body well elevated above the ground (an attitude that a female takes only when she is aggressive) on his three posterior pairs of legs, his head higher that his abdomen, so that the long axis of his body describes an angle of 3o°-40° with the surface of the ground. He waves his palpi upward in the air (i.e., straightening them out before his head) flexes them outward, from one to three times, then draws his body slightly backward and downward, rapidly waving in the air the outstretched palpi and first pair of legs, and spasmodically shaking the whole body with the violence of the movement. The vehemence and to some extent the attitudes reminds one forcibly of a small terrier barking at a cat. The movement of the palpi exhibits most clearly their relatively huge, black terminal joints. When the male is timid, or not very eager, he may wave only his palpi, and these slowly and alternately instead of together. The male repeats these motions several times, usually becoming more ve- hement each time, then moves a step nearer the female, repeats them again, moves nearer again, so that in a short time his out- stretched shaking forelegs come in contact with the female.10 A closely related species, Pardosa saxatilis, raises the forelegs alternately and at the same time wigwags with his jet-black palpi, using them alternately as well. In Pardosa emertoni, the front legs are held up in the air and the palpi are flexed and jerked, and fol- lowed by movements of the abdomen. Pardosa modica makes little 10 T. H. Montgomery, "Studies on the Habits of Spiders, Particularly Those of the Mating Period," Proc. Acad. Nat. Sci., Philadelphia, 1903, pp. 83-4. 82 AMERICAN SPIDERS use of his front legs during the visual courtship but wigwags with his palpi, often standing high on the tips of his tarsi. Within the same group such striking differences in courtship are often found. Lycosa gulosa is a comon grassland spider varying in color from gray to nearly black, and exhibiting only slight differences between the sexes. From its courtship antics it once was given the common name of "purring spider." B. J. Kaston explains: Immediately upon coming in contact with the female, or within three minutes thereof, the male begins to drum his palps rapidly against the floor of the cage. These drumming move- ments are made so rapidly that a distinct purring or humming sound can be heard. The palps are used alternately and are raised only a very short distance during the process. The body is held at an angle so that the posteriar end of the abdomen almost touches the floor. As a consequence when the male be- gins to twitch his abdomen in a vertical plane, the tip strikes the floor. However, I could not detect any sounds made by this part of the body. It is highly probable that the vibrations set up in the substratum by the tapping movements of the palps and abdomen are perceived by the female. This may exert an exciting influence on her in a manner analogous to that which occurs in web-building species, where the male tweaks the threads of the female's snare.11 The male of Schizocosa crassipes has a thick covering of black hairs on the tibiae of the front pair of legs, which are conspicuous epigamic brushes. He extends his long first legs out in front and taps the floor with both about four or five times, simultaneously and in rapid succession. Then the forelegs are raised and the body is elevated high upon the posterior legs, while at the same time the palpi are extended downword to touch the floor below the face. In this position, the brushes on the front tibiae are very conspicu- ously displayed. He advances toward the female with a rhythmi- cally repeated waving of legs, jerking of body, and posturing. A closely allied species, Schizocosa bilineata, bedecked with similar ornaments on the front legs, seems on the other hand to make no 11 B. J. Kaston, "The Senses Involved in the Courtship of Some of the Vagabond Spiders," Entomologica Americana, Vol. XVI (1936) (new series), p. 114. i COURTSHIP AND MATING 83 use of them during mating; in fact, seems to have no visual court- ship at all. The European Pisaura mirabilis is remarkable for its habit of presenting the female with a fly as an inducement to mating. Bris- towe has described the activity in the following manner: A male was given a fly and placed in a box with a female. He proceeded to enwrap the fly with silk, and then walked about with it in a jerky fashion until presently the attention of the female was attracted, and she approached him. He held out the fly to her, and after testing it with her falces, she seized hold of it. The male then crept to a position almost underneath the female, a little to one side, and inserted his right palp. After twenty-five minutes he withdrew his palp and joined the female at the fly. This is rather a remarkable piece of instinct— a car- nivorous creature like a spider deliberately giving up his food as an offering to the female.12 The Peckhams first brought to the attention of naturalists the bizarre courtship antics of the American jumping spiders. The females of this group are for the most part pleasantly colored in grays and browns, while upon the males has been showered an in- finite variety of color and ornament. The chelicerae are enlarged, molded into odd form, and usually colored in iridescent purple, green, or gold. The principal feature of the face is a row of four great pearly-white eyes, and it is embellished above with crests or plumes and overhung with bright hairy fringes. The first legs are wonderfully ornamented with peculiar enlargements of striking colors, and are clothed with fringes of long colored hairs, pendant scales, and enlarged spines. Although less attention has been given to the other legs, they also are sometimes supplied with unusual ornamentation. The Peckhams thought that the male jumpers were much more numerous and that "it was highly improbable that a female ever mates with the first male that comes along. . . . She rejects the ad- vances of one after another; she flies and is pursued; she watches, with great attention, the display of many males, turning her head from side to side as they move back and forth before her; she be- comes so charmed as even to respond with motions of her own body. If we may judge by her attitude, she is observant of every 12 W. S. Bristowe, Proc. Zool. Soc. London, Vol. I (1926), p. 330. 84 AMERICAN SPIDERS posture that the male takes, and appreciative of his every claim of beauty." Whereas we reject the sexual selection of the Peckhams as not truly representing the facts, it must be admitted that the final results are the same. Through the elimination of certain males directly by killing them for food, and indirectly by rejecting them as mates, there is an active female selection. During their antics, the male jumping spiders make every effort to bring into position the striking features of their bodies. Many of them stretch out their front legs and wave them rhythmically and insistently, or take an imposing attitude with arms outstretched like a semaphore. Others lower these legs and keep them motionless so that nothing interferes with the view of the bands and marks on head and clypeus. Some tilt upward to display an iridescent rose or gleaming metallic abdomen. In some the intensity of the dance verges on frenzy, whereas others perform their pantomime with grace and decorum. Some fascinating descriptions are given by the Peckhams. Tutelina elegans is one of the most common eastern American jumping spiders. Both sexes are beautiful. The male is covered with iridescent scales, his general color being green; in the female the coloring is dark, but iridescent, and in certain lights has lovely rosy tints. In the sunlight both shine with the metallic splendor of hum- ming-birds. The male alone has a superciliary fringe of hairs on either side of his head, his first legs being also longer and more adorned than those of his mate. The female is much larger, and her loveliness is accompanied by an extreme irrita- bility of temper which the male seems to regard as a constant menace to his safety, but his eagerness being great, and his manners devoted and tender, he gradually overcomes her opposi- tion. Her change of mood is only brought about after much patient courting on his part. While from three to five inches distant from her, he begins to wave his plumy first legs in a way that reminds one of a windmill. She eyes him fiercely and he keeps at a proper distance for a long time. If he comes close, she dashes at him and he quickly retreats. Sometimes he becomes bolder and when within an inch, pauses, with the first legs out- stretched before him, not raised as is common in other species; the palpi also are held stiffly out in front with the points to- gether. Again she drives him off, and so the play continues. PLATE IX Edwin Way Teak Spider relatives: Harvestmen on aphis-covered rose shoots PLATE X a. Molting. Carapace and chelicerae freed George Elwood Jenks George Elwood Jenks b. Molting. The shed skin TRAP-DOOR SPIDER, Bothriocyrtum californicum Lee Passmore c. Cradle of eggs in burrow TEXT FIG. 2.-COURTSHIP POSTURES OF MALE WOLF AND JUMPING SPIDERS A. Pardosa milvina Hentz. B. Habronattus viridipes Hentz. C. Peckhamia noxi- osa Hentz. D. Hyctia pikei Peckham. E. Peckhamia picata Peckham. F. Euophrys monadnock Emerton. (Redrawn from Kaston, Emerton and Peckham). COURTSHIP AND MATING 85 Now the male grows excited as he approaches her, and while still several inches away, whirls completely around and around; pausing, he runs closer and begins to make his abdomen quiver as he stands on tiptoe in front of her. Prancing from side to side, he grows bolder and bolder, while she seems less fierce, and yielding to the excitement lifts up her magnificently irides- cent abdomen, holding it at one time vertically and at another sideways to him. She no longer rushes at him, but retreats a little as he approaches. At last he comes close to her, lying flat, with his first legs stretched out and quivering. With the tips of his front legs he gently pats her; this seems to arouse the old demon of resistance, and she drives him back. Again and again he pats her with a caressing movement, gradually creeping nearer and nearer, which she now permits without resistance until he crawls over her head to her abdomen, far enough to reach the epigynum with his palpus.13 The largest American jumping spiders are the massive, hairy species of Phidippus (Plates 30, 31 and 32; Plates XXXI and XXXII), which are gaily marked with light spots and often gaudily colored in carmine, orange, and yellow. The face is usually distin- guished by tufts of curled hairs and bands of colored scales and hairs. The elegance of their front legs is especially notable, with long flowing fringes of colored hairs. Some species wave these handsome legs so vigorously that they cross at the tips, but in most instances they are brought to an angle of about forty-five degrees, and, as the male sways toward the female, or approaches her in zigzag fashion, are moved up and down to bring into view the plumes and iridescent plates. Representative of another group very numerous in species is Metaphidippus capitatus. When courting, this species approaches the female rapidly until a couple of inches away, arms extended upward, then stops and drops them down close to the surface. In this position, the face, variegated with snow-white bands and with contrasting gleaming bronze scales, becomes the center of atten- tion. Peckhamia picata (Text Fig. 2, E) is one of the antlike spiders. 13 G. W. and E. G. Peckham, "Observations on Sexual Selection in Spiders of the Family Attidae, Occasional Papers Nat. Hist. Soc., Wisconsin, Vol. i (i) (1889), p. 46. 86 AMERICAN SPIDERS The most important difference in the sexes is the greater thickening of the first legs of the male. These are flattened on the anterior surface and are of a brightly iridescent steel-blue color. Unlike most of the Attid males, this species keeps all his feet on the ground during his courtship; raising himself on the tips of the posterior six, he slightly inclines his head downward by bending his front legs, their convex surface being always turned forward. His abdomen is lifted vertically so that it is at a right angle to the plane of the cephalothorax. In this position he sways from side to side. After a moment, he drops the abdo- men, runs a few steps nearer the female, and then tips his body and begins to sway again. Now he runs in one direction, now in another, pausing every few moments to rock from side to side and to bend his brilliant legs so that she may look full at them.14 The little male of Habrocestum pulex is not so gaily colored as some of his relatives, but he makes up in enthusiasm for his lack of brilliance. His whirling dance has been excellently described by the Peckhams: He saw her as she stood perfectly still, twelve inches away; the glance seemed to excite him and he at once moved toward her; when some four inches from her he stood still and then be- gan the most remarkable performances that an amorous male could offer to an admiring female. She eyes him eagerly, chang- ing her position from time to time so that he might be always in view. He, raising his whole body on one side by straightening out the legs, and lowering it on the other by folding the first two pairs of legs up and under, leaned so far over as to be in danger of losing his balance, which he only maintained by sid- ling rapidly toward the lowered side. The palpus, too, on this side was turned back to correspond to the direction of the legs nearest it. He moved in a semicircle for about two inches and then instantly reversed the position of the legs and circled in the opposite direction, gradually approaching nearer and nearer to the female. Now she dashes toward him, while he, raising his first pair of legs, extends them upward and forward as if to hold her off, but withal slowly retreats. Again and again he circles from side to side, she gazing toward him in a softer mood, "Ibid., p. 43. COURTSHIP AND MATING 87 evidently admiring the grace of his antics. This is repeated until we have counted in circles made by the ardent little male. Now he approaches nearer and nearer and when almost within reach, whirls madly around and around her, she joining and whirling with him in a giddy maze. Again he falls back and resumes his semicircular motions, with his body tilted over; she, all excitement, lowers her head and raises her body so that it is almost vertical; both draw nearer; she moves slowly under him, he crawling over her head, and the mating is accom- plished.15 In the many American species of Habronattus, the front legs and the face are lavished with decoration. The enlarged tibiae of oregonense, the hirsute legs of agilis, the iridescent blue metatarsi of hirsutus, the pink palpi, scarlet clypeus, banded face of other species, are only a few of the expressions of color and ornament in the group. Some of the species have the third legs modified, and among them is viridipes with a strangely formed patella armed with a pale spine and marked with a black spot. During courtship he finds it a difficult task to balance himself while endeavoring to ex- hibit two pirs of legs. When he gets to within an inch of the female, he lifts the first legs nearly at right angles with the body, giving them a bowed position, with the tips approaching each other, so that each leg describes a semicircle, while the palpi are held firmly together in front. Up to this time he has held the body well above the ground, but now he lowers it by spreading out the second and fourth pairs, at the same time bringing the tips of the third pair nearer the body and arching the legs over the posterior part of the cephalothorax in such a way that the proxi- mal ends of the tibiae nearly meet. As he stands in this position, the female, who is watching him eagerly, has the front surface of the apophysis plainly in view over the dorsal surface of the cephalothorax, and face and clypeus are also well exposed. Now he approaches her very slowly, with a sort of creeping move- ment. When almost near enough to touch her, he begins a very complicated movement with the first pair of legs. Directing them obliquely forward, he again and again rotates each leg around an imaginary point just beyond the tip; when they are 88 AMERICAN SPIDERS at the lowest point of the circle, he suddenly snaps the tarsus and metatarsus upward, stiffening and raising the leg and thus exposing more completely its under-surface. While this is going on with the first pair, he is continually jerking the third pair up higher over his back, as though unable to get them into a satis- factory position, and the abdomen is kept twitching.16 Such a display can be carried even farther to include virtually all the legs. Euophrys monadnock is a boreal spider that lives in the moss and lichens of open pine forests, frequently being found in the western mountains. The handsome little male (Text Fig. 2, F) displays the orange femora of his hind legs when he postures before the female. The palpi, jet-black with yellow ends, hung down in front; the first legs, black with pale tips, and fringed with long, thick, pur- plish scales, were thrown diagonally upward; the body was raised high on the tarsi of the second and third pairs, the third being lifted so that the colored femora would be seen over the second, while the legs of the fourth pair were dropped and held at just the angle that brought the femora into view between those of the second and those of the third pair. In this difficult attitude, the spider began to move. There was none of the awk- wardness shown by Pellenes (Habronattus) in trying to keep the third leg in position; indeed, there was no muscular action vis- ible as he glided smoothly back and forth, while the female, turning from side to side, kept him constantly in sight.17 The Web Builders. The spiders in this category are for the most part species that have poor eyesight. Many are confirmed sedentary types or put a considerable reliance on silk, thus effectively obvi- ating a real need for keen vision. In fact, they have extended their snares in ways that carry far beyond the limits of ordinary sight. Through the medium of her web threads, the male is able to per- ceive with reasonable certainty the presence of a female of his spe- cies, and to diagnose her attitude. Courtship among the web spin- ners usually consists of finding out how the land lies, then tele- 16 G. W. and E. G. Peckham, "Additional Observations on Sexual Selection of the Family Attidae," Occasional Papers Nat. Hist. Soc.} Wisconsin, Vol. i (3) (1890), p. 119. 17 G. W. and E. G. Peckham, "Revision of the Attidae of North America," Trans. Wisconsin Acad. Sci., Vol. XVI (1909), p. 360. COURTSHIP AND MATING 89 graphing to the occupant of the web the arrival of the male. In later stages a tactile stroking of the body precedes the coition. The male web spinner has an advantage in that he can approach the female at a distance and is not immediately vulnerable to her attack. A hasty retreat follows notice from the female that he is unwel- come. The whole routine of tweaking the threads, approaching the female on the surface of the web, and further stimulating her at close quarters, constitutes a tactile display of courtship equal in interest to that of the long-sighted spiders. Among the web builders we find some females that are very tolerant of their males and accept their advances eagerly. Some live together quite amicably for weeks, and others are gregarious by habit, spinning huge communal webs in which the sexes live in seeming equality. Withal, there also exist in this group females notorious for their aggressiveness, which are known to destroy their males before or after the mating. Specialization in the orb weavers has taken them along a path where there is a premium for vigorous response to the touching of their snare by any interloper. The spider hurls itself over its web and, by a remarkable exhibition of trapeze artistry, quickly subdues and enswathes its prey. It is there- fore not surprising that so finely trained an aggressor should occa- sionally fail to recognize the advances of a male of her species. The latter is in especial danger if the female is not fully adult, has already mated, or is gravid. After the mating he is in great danger if he tarries too long; it is then that he is most often killed. Among the web spinners are some that are more closely allied to the long-sighted hunters than to their own group, and, except for the silken sheet web over which they run in an upright position, resemble the former in their courtship and mating attitudes. Males of the Agelenidae are in most cases equal to the females in size, and superior in agility. Agelena Pennsylvania a, the commonest grass spider of the naevia group, moves upon the web of the female and signals to her by tapping the silk with his legs and palpi. His ad- vance is usually slow and measured until he is able to touch her with his legs, whereupon he actively seizes her. In most instances resistance is not strong; the male grasps her hind femora in his chelicerae and carries her to the entrance of the silken tunnel, where mating often occurs. He throws her on her side and, his head point- ing in the opposite direction from hers, turns her over and applies the palpi from either side. The tangled-web spinners (family Theridiidae) include many 90 AMERICAN SPIDERS species among which the females show little hostility to the male before mating and enter actively into the preliminary maneuvers. The description of the courtship of Theridion tepidariorum by Montgomery illustrates the general habit of the whole genus. The introductory steps of the mating are as often made by the female as by the male, and she often shows quite an insatiable eagerness, even sometimes leaving food to approach the male. As soon as the male commences to move upon her web she recog- nizes him as a male of her own species, and, when she is eager, commences immediately to signal to him, both spiders being on the lower surface of the web and upside down (the usual posi- tion). The female hangs to the web with the third and fourth pairs of legs, and shakes the longer second and first pairs vigor- ously and spasmodically in the air (when those legs are not at- tached to web lines), otherwise with them she shakes web lines to which they are hooked. This 'signalling' is a sign of eagerness on the part of the female, and so far as I have observed she makes it at no other time than when she is eager and notices the ap- proach of a male of her own species. There are individual dif- ferences in the mode of signalling, as well as differences in accord with the degree of eagerness of the female; sometimes a female signals without moving from her original position, some- times with the signalling she moves by short steps towards the male. When she is not eager she either remains motionless, or else rushes hostilely toward the male as at an object of prey; in both cases the male makes no advances, and when she is mark- edly aggressive he escapes by dropping from the web. The whole attitude of the male is that of combined timidity and eagerness; he is much smaller than the female and upon a foreign web, and usually acts with great caution.18 In this species the female may mate with many males and, except when heavy with eggs, rarely rejects the advances of a suitor. Whereas the male usually leaves hurriedly after mating, in some species of this group he moves to one side of the web, refills his palpi with semen, and returns to mate frequently. Among the sheet weavers of the family Linyphiidae, the male is also a privileged con- sort and is only rarely menaced by an intractable female. 18 T. H. Montgomery, "Studies on the Habits of Spiders, Particularly Those of the Mating Period," Proc. Acad. Nat. Set., Philadelphia, 1903, p. 104. PLATE XI A - •^ * • .--- Passmore Lee Pussmore a. Exposed burrow 4. Male '"Ik* • TOI^:-- Lee Pcasmore c. Cork-door nest held open CALIFORNIA TRAP-DOOR SPIDER, Bothriocyrtum californicum P L A T E XII a. Capturing a ground beetle Walker Van Riper b. Lifting the cork lid CALIFORNIA TRAP-DOOR SPIDER, Bothriocyrtum californicurn COURTSHIP AND MATING 91 The male spider must often coax the female out from her retreat before mating; sometimes he spins a series of lines as a bower in which the pairing can take place. The female of Metepeira laby- rinthea spins a labyrinth of tangled threads behind her orb web and stays in it much of the time. Other orb weavers hide in a leafy retreat near their webs and communicate with the orb by means of a signal line held in the claws of one of the legs. According to G. H. Locket, the English arachnologist to whom we owe much for his keen observations of the web-spinning spiders, the male of Zilla x-notata "climbs to the center of the female's web and usually seizes the line communicating with the female's hiding place with his four front legs. With his back legs he seizes one of the adjacent radii at the centre and starts a series of jerking and plucking move- ments on the communicating line, using himself as a sort of spring at the angle of the radii. If the female does not respond he then usually climbs to her retreat, but returns again after an interplay of legs . . . eventually the female comes out, also making plucking motions, and, after a short interplay of legs, the male begins making thrusts at her epigynum; the palps are then applied alternately." Among the orb weavers we find some species in which the male is a mere pygmy hardly worthy of the female's notice. During the mating season there are often three or four tiny males, only one- fourth the length of the female, hanging in the outskirts of the web of the large orange argiope, Argiope aurantia. They make known their presence by plucking and vibrating the web lines. When advancing toward the female, the male seems to pause and pull at the strands of web, as though to notify her of his approach. When he comes toward her from in front she im- parts a slight motion to the web with her legs, which seems to serve as a warning, as he either moves away or drops out of the web. When he comes from behind she pays no attention to him until he begins to creep on to her body, when she slowly raises one of her long legs and unceremoniously brushes him off.19 No observations have been made of the mating of the bolas spiders of the genus Mastophora. This male is such a tiny creature that he probably has complete immunity from the attack of the female, and clambers over her grotesque body like a tiny parasite. 19Peckham, "Observations on Sexual Selection in Spiders of the Family Attidae," op. cit., p. 55. 92 AMERICAN SPIDERS The need for any courtship in such animated spermatophores should not be very great. THE MATING The transfer of sperm is accomplished in a most amazing manner by means of the palpus and the epigynum. Whereas in most animals the copulatory act contributes little or nothing to knowledge of the group, in spiders the details are of great interest and in many cases of deep significance. During the mating the female becomes quies- cent and remains in a kind of cataleptic state until its termination. In many species, the female first contributes to the mating by align- ing parts of the epigynum so that the corresponding units of the male palpus can be properly oriented. The attitude maintained by the sexes is most constant within the species, and the details some- times give us data on the general position of the spiders of the series. Two principal embraces are found among spiders. In the taran- tulas, the six-eyed spiders, and the aerial web spinners, the male usually approaches the female from in front, and, moving under- neath until his cephalothorax lies beneath her sternum, applies his palpi directly. This is frequently referred to as the Dysdera em- brace. Among the web spiders it is a quite favorable position, since the female hangs inverted below the male and does not greatly menace him. On the other hand, the male wandering spiders that use this type of embrace are in a dangerous situation beneath the jaws of the female. Retreat after the mating, when the female has largely lost her sexual ardor, is likely to be hazardous. In many instances, the males carefully disengage themselves and then leap back and away quickly, showing that they have become conditioned to compensate for a changed attitude in their mates. The second position, the Lycosa embrace, is the one used by the wolf spiders and the running spiders of the higher families. Here the male crawls over the body of the female, and, with head pointed in the opposite direction from hers, reaches around the side of her abdomen to apply a palpus to the epigynum. There is far less danger to the male when he assumes this position, and he is more or less in command of the female until he disengages the palpus and runs away. He serves the right side of her epigynum with his right palpus, and swings around to the other side when using the left palpus. Some of the vagrants have so modified their bodies that it has been necessary to change the type of embrace. Thus, the ab- COURTSHIP AND MATING 93 domen of a female of the stocky crab spiders is often so wide that the male must crawl to a ventral position in order to apply his short palpi. The actual copulation, accomplished by means of the secondary genital structures, consists in the orientation and pressing of the embolus into the atriobursal orifice of the female. Many of the primitive spiders that use the Dysdera embrace apply both the palpi simultaneously to the orifices just beneath the genital furrow. Since both palpi are applied directly from beneath, it follows that the right palpus enters the left orifice, and the left palpus the right orifice, of the female. Other spiders of that series, for example the tarantulas and their allies, apply each palpus alternately, but prob- ably use the same side for insertion as indicated above. In all the higher spiders this situation is reversed, with corresponding palpi serving the corresponding female orifices. This most interesting fact indicates that specialization in the epigynum and palpus has been accompanied by profound changes in the insertion of the embolus. The actual union of the secondary genital structures may be of very brief duration, only a few seconds, or it may be prolonged for several hours. When the organs are highly complicated, the insertion is apparently aided by a preliminary lubrication of the palpus accomplished by drawing it through the chelicerae and is finally consummated only after the manipulation of several different elements. The palpus may be scraped across the epigynum until a spur on the tibia, on the tarsus, or on the bulb itself becomes fixed in a particular groove. Once firmly anchored in this starting point, the palpus swings to assume a position that, with the aid of ridges, grooves, and other processes on the epigynum corresponding to its own outline, makes it possible to guide the embolus to exactly the right point for entering the orifice. At this stage, the bulb of the palpus is still largely in its resting position, lying folded in the cup of the tarsus, and the preliminary contacts serve to hold it firmly in place. Most spiders have at the base of the bulb various thin pouches, or hematodochae, that swell up with the influx of blood until they attain enormous size. This distention causes the entire bulb to turn on its axis, which action forces the embolus into the appropriate opening. The whole embolus (usually a thin spine or heavy spur but often a coil that may consist of several turns) is screwed into the epigynum, following the corresponding tubes in this organ until it reaches the receptaculum seminis. Semen is then pumped into this receptacle by means of a strong blood pressure in 94 AMERICAN SPIDERS the palpus, brought about by contractions of the muscles of the body. The female retains the viable sperms in her receptacles for long periods and dispenses them at the time of egg laying. Although the epigynum has two separate pouches without communication, it is not necessary for the male to supply both of them with sperms to accomplish the impregnation. This usually happens, however, in- asmuch as the male exhausts one palpus and then applies the other one to the other orifice. Polygamy is the rule in spiders, though habits vary. The female and male, if he escapes safely, may pair again. After an initial copu- lation the female may reject forcibly any male that approaches her, or may submit many times to various males, even after her eggs have been laid. In some spiders it is probable that only a single coition occurs, the epigynal openings being blocked with a tough plug following mating. This is especially noticeable in the com- mensal comb-web spiders, Conopistha and Rhomphaea, in which the epigynum is capped with a hard conical cover. In many of the small species of Aranea, notably those of the mineata and juniperi groups, the separate epigynal openings are plugged with a black material so tough that in some instances it has been described as an integral part of the epigynum. In the mated Peucetia, the green lynx spider, the epigynum is covered with a hard blackish layer, probably composed of dried semen and collateral liquid, and there is usually present a small process from the male palpus, broken off during the mating— a fact which aids greatly in associating the proper male and female when there are more than one species of Peucetia in a particular region. PALPUS AND EPIGYNUM The pedipalps are the second pair of appendages of the head and lie on each side of the mouth, being inserted behind the chelicerae. They are six-segmented organs consisting, from base outward, of coxa, trochanter, femur, patella, tibia, and tarsus, and thus lacking the additional metatarsal segment present in the legs. The basal segment is the coxa, and it usually bears a conspicuous lobe, the endite or maxilla, that lies at the side of the labium and serves as a cutting and crushing instrument while feeding. The remainder of the pedipalp is the leglike palpus, whose tarsus is ordinarily armed PLATE 13 Walker Van Riper, Colorado Museum of Natural History a. Portrait of a tarantula, Aphonopelma &€>-•.* Walker Van Riper, Colorado Museum of Natural History b. Side view of tarantula, Aphonopelma TARANTULAS PLATE 14 J. M. Hollisler a. Banded Argiope, Argiope trifasciata, in web J. A/. Hollisler b. Spiny-bodied spider, Gasteracantha cancriformis, on leaf ORB WEAVERS COURTSHIP AND MATING 95 in females with a single terminal claw, while in the males it is en- larged and transformed into a copulatory organ. The spider's palpus is undoubtedly the most unusual intromittent organ that has been developed in any group of animals; its parallel is found only in some crustaceans, in dragonflies, and in the rare arachnids of the order Ricinulei. The whole process of its employ- ment is a complicated one, requiring detailed, elaborate acts and routines before it is successful. What do we know about the devel- opment of this curious mode of copulation? There is no recapitu- lation of its origin and evolution in the lives of spiders themselves. Nevertheless, nature usually accomplishes new things by small steps and leaves behind traces of the path that has been followed, omitting only a few missing links to be filled in by speculation. T. H. Mont- gomery and various other araneologists have attempted to outline the spider's path, and they have all followed the same line of rea- soning. Mating among the earliest spiders or their precursors must have been by means of direct contact of the genital openings. In insects and in some arachnids as well, the male intromittent organ is directly connected with the vas deferens, and through this tube courses the products of the testes. The scorpions appress their abdomens close together, effecting the transfer by means of an eversible copulatory organ. This same habit is found in the harvestmen, but the intro- mittent organ is usually a long tube that conveys the semen. The prime step toward araneid copulation is that of voiding a sperma- tophore and then transferring it by means of an appendage, thus doing away completely with direct copulation. The male solpugid,. one of the most primitive of all arachnids, seizes the female, and,, by pinching her abdomen, causes her to fall into a state of torpor,, whereupon he ejaculates and transfers the semen to her receptacle with his chelicerae. Among the pseudoscorpions, the male grasps, the hands of the female in his and pulls her back and forth in a courtship dance, displaying at the same time the ram's horn organs at the base of his abdomen. When the female is sufficiently stimu- lated and responds with the necessary dancing movements, the male lets go her hands and extrudes a globule of semen or sperma- tophore that is attached to the floor by a thread and stands free on this line. At just the right moment, when the female dances over the spermatophore, the male grasps her genital cleft with his stout front legs and forces the drop into the aperture. Comparable habits. 96 AMERICAN SPIDERS are known among the mites, where the chelicerae or legs are used to transfer the semen. The final step in araneid copulation is the modification of the appendage into an organ where the sperm can be stored some time before pairing. Having the semen secure in a reservoir at the end of the palpus does away with the need of ejaculating it during or immediately preceding the mating, and lessens the risk of losing the female during such a preliminary routine. The intermediate stages of the araneid mode of copulation have been dropped out com- pletely, and do not even persist in some form in the memory of the race. What we see is the culmination of the whole process, something unbelievably complicated, the proper performance of which is part of the instinctive makeup of the male spider. The palpus of the male only gradually developed into the com- plicated organ we now observe. At first it resembled that of the female, and was armed at the end with a single tarsal claw that picked up the spermatophore and pressed it into the female vulva. Gradually the claw became transformed into a cup-shaped recepta- cle, from which the liquid was less easily lost, and finally the cup was closed at the end until only a small opening remained through which to take in and drain out the semen. This receptacle is the all-important element of the palpus, and in its simplest form is made of three more or less well-defined parts: a basal expanded portion termed the fundus, a coiled intermediate tube called the reservoir, and the delicate terminal ejaculatory duct. We see this elementary receptable in the palpi of many spiders where it has remained very simple, and can still discern it as the prime element of those in which the structure has become greatly elaborated. At first, the receptaculum seminis was appended to the tarsus as a simple extension, but in this position it was quite liable to be broken or injured in some way. Specialization has proceeded to protect it and its delicate terminal duct, and to make it more effec- tive as an intromittent organ. Around it has been developed a pro- tective cover called the bulb. The tarsus itself has been excavated to form a receptacle in which the whole organ can lie when at rest. Muscles and blood pouches have been evolved to make possible the ejection of the semen. On the bulb itself have arisen processes that are used to orient the parts of the palpus in relation to those of the epigynum, and apophyses on the tibial and other palpal segments to act in a similar manner. As is true for the male, the gonads of the female are hidden deep COURTSHIP AND MATING 97 in the abdomen. From the two ovaries come off oviducts that join to form a single tube, the uterus, which opens externally through a transverse slit at the middle of the epigastric furrow near the base of the abdomen. It is probable that in ancient spiders the sperma- tophore of the male was pressed into the small opening, or into a pouch formed in front where it was to be stored. There are still many spiders having only simple paired receptacles just in front of the genital furrow, and no external evidence of the organ we know as the epigynum. All the tarantulas, and the many true spiders with generalized palpi (the Haplogynae), belong in this series, possessing relatively simple male and female genitalia. The males introduce their palpi, frequently simultaneously, into the transverse genital opening and into the receptacles— known as spermathecae— which serve as storage vessels for the sperms. Only the terminal part of the organ, the embolus, is pressed into the spermathecae. The epigynum of the female is composed of two essentially sym- metrical independent units, each of which serves as a sheath for the male embolus of its particular side. There is a very close corre- spondence between the physical proportions of the duct leading to the female spermatheca, and the embolus of the male, a natural result of parallel evolution. In all the higher spiders there exists a pair of outside openings into which each male embolus can be in- serted without gaining entrance through the medial genital pore within the body itself. Much of the perfection and elaboration in the palpi and epigyna must be attributed to this new position of the orifices, which makes possible the adoption of a different mode of pairing, in which the male is less vulnerable. The external epigynum has become specialized in various ways. In many spiders there is an atrium, surrounded by a distinct rim, within which lie the orifices. These latter are separated by median ridges that guide the embolus into its particular channel. At the front or behind may be a hood or tubercle articulating with an apophysis of the palpus. A conspicuous finger often overhangs the atrium and serves to fix the palpus in just the right position to make the pairing possible. It should be noted that not every apophysis of the male palpus has been developed to fit a corresponding depression in the female epigynum. As in all structures of animals, the ortho- genetic tendency to become more elaborate may go far beyond the needs of the animal. Many of the spurs and strange projections in complicated palpi may prove only useless luxuries that contribute nothing to mating. 98 AMERICAN SPIDERS The very close correspondence between the male and female genitalia of insects ted Leon Dufour to the formulation of the so- called "lock and key" principle. It was his belief that the crossing of species was impossible for physical reasons, and that the male organ could not be introduced into that of a strange female because of differences in length, shape, and size. The female organ was re- garded as an unyielding lock that could be opened only by a key that corresponded exactly with its form. Whereas we must reject the theory that these organs are adaptations that exclude the cross- ing of species, and instead assign that function to fundamental in- stinctive patterns probably based on chemical stimuli, it must be admitted that in spiders the differences between the genitalia of allied groups are usually sufficiently great to make pairing impos- sible—in effect a "lock and key" presenting an impassable barrier to all but the most closely related species. It must be kept in mind that the secondary genitalia of spiders are extremely ancient organs probably fully evolved long before the late Paleozoic Era, where we find fossil spiders. Both primitive true spiders and living tarantulas, discretely separated even at that time, have similar palpi, indicating that the general features of their organs antedate the separation of the two suborders. It is little wonder then that in the palpi and epigyna are clues to the general phylogeny of the whole group. These organs have undergone changes corresponding closely with the specialization of spiders themselves. Indeed, sexuality and the araneid mode of copulation are adaptations that have probably contributed more to spider evo- lution than have any other features. CHAPTER VI The Evolution of Spiders OPIDERS OWE MUCH TO THE PAST. The proud jumping spider of today, attired in flowing robes of ermine and crimson and with great smoky eyes intently following every moment of a gleaming bluebottle fly, bears little resemblance to its reserved, myopic forebears. The sedate orb weaver, hanging from a web of wondrous design, has come a long way from the clumsy land creature that first attempted to climb into the shrubs. So changed are many spiders that we can scarcely discern in their bodies any clues pointing to their origin. From fossil evidence we know that spiders are ancient creatures, and that they were confirmed land animals before the vertebrates had got free of the bondage of aquatic life. A large part of their evolution must have been undergone during the Devonian Period, which has left one record of an enigmatic spider, Paleocteniza, from the Rhynie Chert of Scotland, occurring with mites and numerous excellently preserved arachnids of the extinct Anthraco?narti. Splen- did fossils come from the coal measures of the Carboniferous Era, in both Europe and the United States, revealing that at that time highly developed, typical spiders were already in existence. Much remains to be learned of earlier araneids, and of the arachnid group that gave rise to them, since we have no evidence to show that spiders have been derived from any other living or extinct group of arachnids. Nor do we have any conclusive evidence that the arachnids evolved from any particular arthropod group. The classical theory of Ray Lancaster, which postulates the trilobites as the ancient group from which have been derived scorpions and typical arachnids on one hand, and eurypterids and king crabs on the other, has been se- verely criticized. More recent evidence, however, strengthens this general thesis and points to the derivation of these diverse arachnid groups from relatives of the conservative trilobites living in the 99 ioo AMERICAN SPIDERS Cambrian seas. An alternate theory would have the arachnids de- rived from some land creature, similar perhaps to the sluglike Peri- patus, but of which at present there exists no record. The phylogeny of spiders has long been the subject of much speculation, and there is still no general agreement as to the funda- mental paths that were followed. This volume attempts to lay down only the broad features of spider evolution, and acknowledges the inclusion of much speculative matter. The phylogeny of any group of animals can be postulated by means of the fossil record, and also by aids from taxonomic classifi- cations, which are frequently indicative of the racial history of a group. The ancestral stock from which come all major spider groups originated some time before the Carboniferous Era. These creatures probably bore a close resemblance to the spiders fossilized in the coal measures, with abdomens encased in hardened plates— wide ter- gites above, sternites below, and hard, narrow pieces (pleurites) on the sides. Four pairs of similar fingerlike appendages were present beneath the abdomen at about the middle. Just in front of these was a pair of spiracular openings leading to book lungs, and a second similar pair was present farther forward, at the base of the abdomen. Inside the abdomen was an elongate heart into which opened five, six, seven, or even more pairs of ostia, each pair representing a seg- ment and those at the rear much reduced in size. Thereafter, the tendency was to increase the size of the organs in the anterior seg- ments and gradually to suppress the posterior ones, resulting in the loss of some of the ostia and the supporting muscular systems. The gradual reduction and loss of the units of internal segmentation were matched in the external plates, which resulted in an actual migration forward toward the spinnerets of the anal tubercle. The cephalothorax of the ancestral spider was relatively long as compared with its width, and was marked by a longitudinal median groove. At the front end were eight eyes set close together on a low tubercle. The legs were of moderate length, quite stout, and each tarsus had at its tip three tarsal claws, the outer paired ones relatively long and smooth, the inner unpaired one short and only slightly curved. The chelicerae were large, set parallel to the long axis of the body, with robust fangs. The gland in the basal segment secreted a weak poison, largely unnecessary since to subdue prey reliance was placed mainly on the strong legs and sharp fangs. The earliest spiders were cautious hunters that groped about on THE EVOLUTION OF SPIDERS 101 the ground and made little effort to establish a permanent station of refuge. Food perception was accomplished by sensory leg hairs which tested the terrain, for their small eyes were useful only to distinguish light from darkness. These sluggish prototypes lived a timeless life of leisure on the tangled jungle floor of their humid swampland. Only during molting and egg laying was it desirable to be concealed from wandering predators, and from less worthy adversaries that under those trying circumstances might do injury to the eggs or to the spiders themselves. The first step toward a life of dependence on silk was the coating of the eggs with excre- tory material from the abdomen, voided by coxal glands that opened through the abdominal appendages. As the product of the glands became more suitable for use as a gluing and covering agent, and the spinnerets more adept in their application of the gummy liquid, greater possibilities for the use of the crude silk opened on all sides. These early spiders were perennials. Each female produced and cared for many egg masses during her life with varying degrees of efficiency and success. Those survived that were more adequately protected by a silken cover, and guarded in long vigils by the mothers, whose regard for the safety of the egg mass was being tried and modified by an increasingly hostile and enterprising band of predators. By the late Paleozoic, the two principal groups of spiders known today had been developed: the My galomorphae , or tarantulas and their allies; and the Araneomorphae, the true spiders. Discretely separated even in the coal measures, these two lines have grown up side by side. In many respects their accomplishments have paralleled each other, a natural development since both were originally en- dowed with similar equipment and potentialities. However, for various reasons, the true spiders surpassed the tarantulas during the Tertiary and became the dominant group. THE TARANTULAS Our first sight of the typical mygalomorph spider is in the coal measures, where we find it little changed from the ancestral spider that preceded it. During the Paleozoic Era, when much of North America was a dismal, swampy area covered by great forests of strange plants and trees, there lived in the region of modern Illinois primitive spiders whose abdomens were armored with hardened loz AMERICAN SPIDERS plates. So familiar is their outline that we immediately associate them with spiders visible about us, and discern a close resemblance to the liphistiids and the trap-door spiders. During the same era, this early mygalomorph was also found in Europe, more numerous in species and so much more diversified that the imprints seem to belong to several distinct types. All have well-marked tergites. The Illinois spiders from the Pennsylvanian shales of Mazon Creek are placed in the genus Arthrolycosa, and in the family Arthrolycosidae, and they are remarkably like the modern species of the family Liphistiidae. However, nothing is known of their spinnerets, claws, sternum, or of other features largely used in clas- sification. After a brief glimpse of them in the coal measures of the northern hemisphere, we lose sight of them completely and can only speculate on their subsequent history. From creatures like Arthrolycosa and its European cousins has been developed all the assemblage of modern spiders known as liphistiids, trap-door spiders, funnel-web tarantulas, and typical ta- rantulas; in short, all of the My galomorphae in the broad sense. If we agree with Eugene Simon, the master arachnologist, that the liphistiids are only primitive members of the mygalomorph spiders, we have no difficulty in accounting for the restricted, more typical recent members. The insistence of many specialists that the myga- lomorph spider of the Paleozoic completely lacked dorsal segmen- tation of the abdomen is unreasonable. The tarantulas are present in the Paleozoic with plates on the back of the abdomen; and many of them have retained well-marked evidences of dorsal segmentation through three or four hundred million years until the present time. THE TRUE SPIDERS Certain shadowy forms from the Carboniferous Era, contempo- raries of the oldest tarantulas, have been assigned with some con- fidence to the Araneomorphae, or true spiders. They appear to lack hard plates on the abdomen, and to assume— in a vague way at least— the form of some of the highest spiders. In what ways do these emergent creatures, from which is derived the vast array of mod- ern true spiders, differ from the Paleozoic tarantulas? How did the branches separate? The fundamental change may well have been one of behavior, a change in habit or attitude rather than a physical alteration. In some THE EVOLUTION OF SPIDERS 103 way it is related to their greater use of silk, to their more expert spinning, and to the retention of the anterior median spinnerets as functional organs until the principal lines of true spiders were well established. Associated with this divergence from the tarantulas was the gradual change in position of the chelicerae. In modern representatives these are now twisted at right angles from the long axis, with the fangs pointing toward each other. Just what advan- tage this development brought is not completely clear, but it may be that the improving eyes and the new chelicerae could be used together to subdue insects more effectively. Cutting edges were being developed on the coxae of the palpi to aid in crushing the body of the prey. The venom was becoming more potent and the voluminous glands were pressing beyond the limits of the cheliceral segments into the head itself. Loss of the heavy abdominal plates was another consequence of the change in life. This armor disappeared gradually and still is vaguely indicated in a very few modern true spiders. P ale o diet yna, a spider from Baltic amber, retains conspicuous plates; and this slow divestiture suggests the possibility of finding many more fossil true spiders retaining dorsal plates. The course of true-spider evolution has been charted largely by silk spinning. The Araneoworphae began their history with the same equipment as the parent tarantula group— eight functional spinnerets of nearly equal size. But whereas the tarantulas were content to spin in a modest way, the true spiders began to use silk more often and with greater efficiency. Since the lateral spinnerets undoubt- edly were bisegmented at an early date, and had the advantage of greater length and strategic position, it was natural that these should be developed and improved at the expense of the unisegmented median pairs. The great reduction in size and early loss of both anterior median and lateral spinnerets in all but a few relict myg- alomorph spiders reflect their failure as spinners. The true spiders, on the other hand, retained all these spinnerets for a long period, and some still keep the anterior median pair. In this connection it is worth noting that the metatarsal comb— the calamistrum— used to brush across the spinning field of the median spinnerets, was in all likelihood an early invention, and that all true spiders once spun cribellate threads. The retention of the anterior lateral spinnerets as prime spinning organs, probably made possible by persistent use of the incipient cribellum, was the key to true-spider superiority, io4 AMERICAN SPIDERS and actually caused the divergence of the true spiders from the parent line. It was inevitable that, in addition to the formal silken covering over the egg mass, many threads would be scattered more or less haphazardly from this spinning center. Such wild lines were in- strumental in giving to the mother spider another advantage in her efforts to guard the eggs, communicating the approach of an inter- loper by vibrations on the threads. The range of touch perception was thus in one step expanded far beyond mere contact with the sensory hairs on legs or body; the deadly predator or the blunder- ing insect often became the prey of the vigilant spider. In this two- dimensional maze of threads, with the egg sac as central theme, was the germ of all the webs that have made the true spider dominant. The stringing-out of silken lines continued during the whole life of the spider, as well as at the egg laying, and has continued to the present time as the dragline habit of modern spiders. With a secure line attached to the spinnerets, the spider could now venture upon precipitous surfaces with a certainty of quick recovery from falls. Since the dragline of true spiders is ordinarily spun through the anterior lateral pair, the tarantulas, in suppressing these spinnerets, virtually precluded the future possibility of becoming aerial spiders. The lifeline of the whole group of true spiders became their silken threads, and those that refused to accept subservience to this material died out. Every spider became sedentary to a degree, and none has been able to divest itself completely from silk since those early days. Each major group of spiders diverged from the others with essentially the same type of spinning equipment, and with a well-founded instinctive knowledge of silk spinning. In each of the lines similar types of webs and traps for the capture of insects have been evolved separately. In one group the anterior median spinnerets have been perpet- uated in a modified form as the cribellum. These creatures come down to modern times in a more or less homogeneous line as the "cribellate" spiders (p. 137). The whole series probably diverged quite early from the main stem, and, although their physical fea- tures mark them as a more generalized group, they have done re- markable things with their heritage. All the remaining true spiders lost the anterior median spinnerets, but in most of them vestigial evidences can still be observed. During the early history of the ecribellate true spiders, a trend- already running a similar course among the cribellate types— toward THE EVOLUTION OF SPIDERS 105 the simplification of various organs was operative. The mutations began at different times and progressed at different rates, so that in modern types generalized features sometimes exist side by side with profound specializations. The tendency has been to simplify the fundamental systems, to make fewer segments and functional units (such as book lungs, tracheae, ostia, spinnerets) do the work of the greater ancestral number. The abdomen was developing into a highly developed center for silk spinning, and in most lines tended to become shorter; in some, globose. The spinnerets gradually attained a position at the tip of the abdomen, near the anal tubercle itself, indicating the virtual re- duction of the abdominal segments to four. Some of the spinnerets later became elongated and modified to perform special types of weaving, and others became so reduced in size that in certain cases only the anterior lateral pair remain as functional spinning organs. A notable achievement was the transformation of the hind pair of book lungs into a pair of tracheal tubes soon after separation from the cribellate line; this development was followed in most spiders by fusion of the openings into a single tracheal spiracle. In a few lines the front pair of book lungs was also converted into tracheae. The number of ostia in the heart was reduced from four to three pairs, in some species even to two pairs, and the remaining ostia assumed the function of the lost members. Changes of many kinds were also taking place in the cephalo- thorax and its appendages. Especially notable was the migration of the eyes from the original local center at the front edge of the carapace, to the sides and to other positions of greater advantage. The anterior median pair was lost early by a whole group of species that persists until the present time as six-eyed spiders, and whose other characteristics indicate that they are among the most gen- eralized ecribellate true spiders. Other types enlarged their eyes, and, with appropriate changes in the legs and body, came to place considerable reliance on sight as an aid in gaining a livelihood. The early ecribellate spiders were at first terrestrial types that stalked over the soil and low vegetation in an upright position, trail- ing their dragline threads behind. Some of the lighter ones began essaying trips into the herbs and shrubs, and learned to use their claws to climb from twig to twig, hanging back-downward from their silken lines. The third dimension was becoming a spacious reality to these extroverts, and with its spaciousness came complete freedom from attack by ground creatures. The egg sac was in- 106 AMERICAN SPIDERS stalled in the center of the tangle of threads, completely safe from flying predators, which could not reach it without becoming en- meshed in the lines. And from entangled insects of many kinds the spider was securing its food. The aerial web spinners (p. 157) be- came specialists for life on silken lines, modifying the unpaired claws of the tarsi into effective hooks. Many spiders remained creatures of the soil, and for running or climbing made little or no use of the unpaired claw. Some of these hunters (p. 193) lost the unpaired claw, developing instead adhesive claw tufts that allow great ease of climbing. CHAPTER VII The Tarantulas TH HE NAMES OF TARANTULA, TRAP- door spider, purse-web spider, and liphistiid bring to mind some of the most famous of all spiders— spiders that rival in size the largest land invertebrates, spiders that have become renowned for their wonderful burrows and handiwork. All are four-lunged spiders belonging to the suborder Mygalomorphae; they are often referred to as mygales but in this book are collectively known as "taran- tulas" or mygalomorph spiders in contrast to the "true spiders" of the suborder Araneomorphae. The mygalomorph spiders are more generalized than the true spiders and ancestral to them. As a group they are longevous, all living more than a single year and some of them attaining great age— as age is measured in invertebrates: up to or even exceeding twenty-five years. They are large, probably averaging more than an inch in length as compared with less than one-fourth that size for the true spiders. Some of the typical tarantulas attain a body length of three and one-half inches; at the other end of the scale, the pygmies, the tunnel and sheet weavers of the genus Micro- hexura, are one-eighth inch long. Along with great size the myg- alomorphs perhaps retain as a consequence the second pair of book lungs and other generalized features correlated with their primitive station among spiders as a whole. Although it must be conceded that the true spiders have attained a higher degree of development— as evidenced by their greater numbers, variety of structure, and multiplicity of habit— the taran- tulas should not be thought of as vastly inferior. They have become notably specialized in their own way, and in instinctive behavior have nearly kept pace with their cousins. The most important single character that distinguishes the myg- alomorph spiders is the articulation of their chelicerae— termed 107 io8 AMERICAN SPIDERS paraxial as contrasted with the diaxial position of true spiders— and other details of the mouth parts. The chelicerae (Plate XV) are robust and two-segmented, as usual, but with their long axis par- allel to that of the body, and with movement in a vertical plane. As befits these powerful spiders, the fang of the chelicera is a stout, curved weapon. In order to drive the fang into the victim, the body must be elevated. These creatures strike with great speed, but be- cause of their poor eyesight and the necessity for waste motions, their method is probably inferior to that of the true spiders. When confronted by man or any creature outside its normal experience, the tarantula throws itself back and maintains its body in a position of readiness to strike. This is a defensive attitude, but also one fa- vorable for attack. The venom glands of the mygalomorph are entirely contained within the basal segment of the chelicera. Since its offensive needs are met by a powerful body and robust jaws, the necessity for great quantities of potent venom is minimized. In most tarantulas the coxa of the palpi also lacks the endite or maxilla, an expansive lobe used in crushing and cutting the prey. All the Mygalomorphae have two pairs of book lungs, clearly visible on the ventral surface of the abdomen and notable for their large size. Only one family of true spiders, the Hypochilidae, has retained this four-lunged condition, and they are the most gen- eralized of all true spiders in many other respects as well. A moderate number of mygalomorph spiders range up into the temperate zones, but the group is essentially tropical and subtropical in distribution, about fifteen hundred species being known from these zones all around the world. During the Paleozoic Era, their ancestors dwelt in the swampy, humid forests that became the coal measures of the United States and Europe. No tarantulas are known from the Mesozoic, but we can be sure that they were well rep- resented, and perhaps at that time equaled the true spiders in num- bers and variety. Because of their secretive habits, which have resulted in a meager fossil record, few Cenozoic mygalomorphs are known; small numbers have been found in the Baltic amber of the Oligocene, and in the Oligocene shales of the Florissant formation. At some time during the early history of the Mygalomorphae the line split into two principal branches, which have descended to us side by side as our modern fauna. On the one hand are the typical tarantulas and the trap-door spiders; they represent the largest and the best-known series. The second group is somewhat inferior in THE TARANTULAS 109 physical equipment (if we measure this in terms of the degree of change from ancestors), and has come down as a reminder of what most of the mygalomorph types were like during the Mesozoic. These latter we refer to as the atypical tarantulas. THE TYPICAL TARANTULAS In this series, which includes the tarantulas, the sheet-web ta- rantulas, and the true trap-door spiders, there is no visual evidence of dorsal segmentation of the abdomen. The maxillary lobes are not at all developed in the American species, but in some exotic forms a small angled spur or a well-developed process may be present. Nearly all have but four spinnerets, the hind lateral and median pairs; these are located close in front of the anal tubercle. The commonly associated characteristics of tarantulas— large size and hairy covering— should not mislead one in identifying members of this group. Many are relatively small in stature. Only the wander- ing hunters, the true tarantulas, are thickly clothed with velvety wool and long silken hairs; others appear quite naked by compar- ison. A few of the mygalomorphs have become vagrant, but none has attained the degree of freedom enjoyed by certain true spiders. Failure to improve vision has resulted in the development of very few accomplished runners, jumpers, or climbers, and none of these tarantulas has become dependent on silk as have the aerial true spiders. Their reliance on touch is perhaps even stronger than in the araneomorphs; the hairy covering of the vagrants, for example, serves admirably to make them aware of the presence of their prey. The typical tarantulas have been most successful in living a secretive life hidden in the ground, with the consequence that many have become specialists in subterranean existence. Their general makeup fits them eminently for a successful life in tropical regions, where competition is not so keen. Few Americans realize that in the southern portion of the United States exists a rich and varied fauna of mygalomorph spiders, eighty or ninety species including many with curious habits. Trap-Door Spiders. Many spiders tunnel into the soil, but the true trap-door spiders of the family Ctenizidae are the most accom- plished burrowers and the most gifted artisans. They and their relatives can claim to be the inventors of that superb mechanism to no AMERICAN SPIDERS ensure privacy, the trap door, for they represent a stock that was probably capping burrows with doors long before many true-spider emulators were evolved. The first description of this interesting device was given by Patrick Browne, who in 1756 illustrated the nest of a West Indian species in his Civil and Natural History of Jamaica. A few years later the nests of Nemesia were described from France, being likened to "little rabbit burrows lined with silk and closed with a tightly fitting movable door." Although trap- door spider nests continued to attract popular attention thereafter, it was not until 1873, when J. Traherne Moggridge published his careful studies on the habits of these animals, that any comprehen- sive treatment was accorded them. Moggridge was able to distinguish four distinct types of nests among the species he studied. The first was a simple tube, a cylin- der closed with a thick, beveled door, which he termed the "cork door"; the second was a simple tube closed with a thin or "wafer" door; and the third type was a simple tube with a thin outer door and a second door part way down. Moggridge's fourth classifica- tion was the most complicated: a nest capped on the outside by a thin door, and having an oblique side tunnel, connected with the main tube, at the entrance of which was a trap door. Several other types of nests have since been discovered in various parts of the world, some of them much more complicated than those described by Moggridge. Furthermore, the distinction between the cork door and the wafer door, while valid enough in the extremes of each type, gradually disappears as we examine long series of inter- graded nests. The true trap-door spiders have developed a comblike rake of large spines on the margins of their chelicerae, and this they employ as a digging instrument. With its aid they are able to cut and scrape away small particles of earth, which they mold into balls and carry outside the burrow. They waterproof the walls of the tube by applying a coating of saliva and earth, so that the surface becomes smooth and firm. Then they apply a silken lining of variable thick- ness and extent, in some cases not fully coating the burrow, while in others covering the whole tunnel with a thick fabric. When the maturing spider outgrows its burrow, it enlarges the domicile by cutting and scraping off bits of earth with its rake and carrying them away from the site. Rocks embedded in the soil may oblige the spider to pursue a tortuous course, or to dig a new tunnel in a more favorable situation. It rarely deserts its burrow PLATE 15 Purse web of Aiypus abboti against tree J. M. Hollister PLATE l6 a. Door open Robert E. Ball Robert E. Ball b. Door half open Robert E. Ball c. Door closed BURROW OF FOLDING-DOOR TARANTULA, AntroJ THE TARANTULAS in voluntarily. When forcibly removed, it will accept the unoccupied tunnel of another spider, or a cavity especially made for it, and proceeds to remodel this in a characteristic way to suit the pattern of previous homes. Although spiders of many other families burrow, the trap-door mygalmorphs have far outstripped them in the excellence of their tunneling. They have become specialists that dig with better in- struments, line with greater care, and are the originators of the intriguing practice of capping the burrow with a perfect lid. While this trap door is not a unique accomplishment of these spiders, having been developed independently by several other groups, it bespeaks a mastery not closely approached by any emulator. The typical burrow is spacious enough in part of its length to allow the spider to reverse position at will. Within its confines the spider finds a haven until violent or natural death. What are the advantages of this abode, which has become such a dominant ele- ment in the lives of these spiders? In the first place, it is the property of a single, unsocial individual and can become, with the passage of time, more and more adequately coated with silk, more and more familiar in its every part, and thus increasingly acceptable to the spider. It is a retreat from the rays of the sun, the extreme heat of which is shunned by nocturnal and diurnal forms alike. Its hinged lid, which can be opened or closed at will, prevents rain and surface water from entering, thus keeping the nest drier than surface situations. Since all the burrowing spiders live more than a single year, the tunnel serves to temper the extremes of inclement weather over long periods. The tube beneath the surface is cooler during the summer heat, and somewhat warmer during the extreme winter cold. Relatively inconspicuous because of its location on the surface of the ground, the burrow opening may be made even more difficult to discern through the efforts of the spider. During the hottest part of the summer, when inimical parasitic wasps are pres- ent in maximum number, the opening may be closed tightly with earth and silk. Mosses, leaves, sticks, and other debris can be placed to advantage on the lid and around the entrance, the result— to human eyes at least— hinting of camouflage. When in active use, the burrow can serve as an ambush from which the spider rushes out to seize its prey; and once an insect is caught, the nest becomes in most cases the dining room. At the proper season the burrow may also serve as a mating chamber, the eggs being laid and en- closed in their sac within its confines. Later it becomes the home ii2 AMERICAN SPIDERS of the young spiderlings, often for many weeks after their emer- gence from the egg sac. The opening to the surface is the spider's only contact with the outside. It is the vulnerable element in the circumscribed abode, but at the same time it allows the creature to be menaced from only one direction. On the surface, an inferior sensory equipment places the trap-door spider at great disadvantage in combat with its specialized enemies. Within the burrow, it faces the enemy pro- tected by a silk door, and should that be torn away, it still has a favorable situation for the use of its strong jaws. While demands for privacy have probably inspired the perfec- tion of the underground castle of the trap-door spider, it is more intriguing to think of the domicile in terms of response to the rav- ages of some arch enemy. By far the most fearsome assailant is the spider wasp, a common name for various species of Pompilidae, which are exclusively spider predators. Other enemies may wreak their toll in an insidious way and possibly destroy more individuals than does the wasp, but this gleaming tyrant is a predator of the first magnitude whose prey is the large, adult spider and whose victory is won in hand-to-hand struggle. Actively foraging over the soil, unerringly directed by a sense not conditioned by previous experience, the wasp arrives at the trap door, beneath which sits the prospective victim— possibly aware, through its delicate tactile sense, of the presence of an in- truder. If unprepared, or if its resistance is finally broken down, the spider quickly finds itself confronted by an enemy that has lifted the trap door or gnawed through it and entered the spacious bur- row. The struggle that ensues is not a battle of giants. It is a very unequal one from which the wasp almost always emerges the victor. Swift and sure in movement, liberally endowed with fine sensory equipment, and armed with a deadly sting, the wasp confidently assails a larger creature fighting on a prepared battleground in the deep recesses of its burrow. After a brief struggle the wasp para- lyzes the spider with venom from its fiery sting, whereupon it proceeds to deposit on the spider's abdomen an egg, from which will hatch a voracious larva. Doomed to lie helpless while furnishing fresh food for the larva, virtually dead if not actually so, the once mighty spider finds its castle converted into a crypt. Industrial skill has failed to make the burrow impregnable to its most formid- able enemy. During the growing period, when the spider is remodeling and THE TARANTULAS 113 strengthening its closed tube, it is less subject to the attacks of marauding wasps that, in order to satisfy the food requirements of their offspring, pass up the smaller burrows in favor of mature or nearly mature prey. We pass now to consideration of the three better-known types of trap-door spiders found in the United States. The first of these constructs the classical type of nest that Moggridge called the "cork nest." The most familiar domicile of this type is made by Bothriocyrtum calif ornicum, the common trap-door spider of south- ern California. Examples of this nest (Plate XI) are to be found in many collections, and may even be purchased from various biologi- cal supply houses. It is the typical nest illustrated in many works on natural history. Another group of spiders that is widely dis- tributed across the southern United States, the genus Pachylomerus, also makes this type of nest. These spiders are very handsome animals, with a nearly oval, black, shining cephalothorax and legs, and a dusky abdomen. The cork nest (Text Fig. 3, A) is a simple tube without side branches, lined completely with silk. Ordinarily the burrows are shallow, from five to eight inches in depth, with a diameter essen- tially the same throughout and great enough, especially near the entrance, to permit the spider to turn around. The distinctive fea- ture of this nest is the door (Plate XI). It is made of layers of earth and silk, and is so constructed that it fits perfectly and tightly closes the mouth of the tube "much as a cork closes the neck of a bottle"— so Moggridge described it. The cork door cannot stand open; it falls and closes of its own weight, and the tube mouth is beveled to receive it. In West Florida Pachylomerus audoumi digs its burrows in the sides of steep, stream-cut banks in moist and shady ravines. Ari- zonan and Mexican Pachylomerus favor open spaces in the sun- baked creosote-bush deserts. In southern California Bothriocyrtum californicum makes its tunnels on sunny hillsides that in early summer bear a thick covering of native grasses. The spiders that build the cork nest are plump animals (Plates XII and XIV) with rather short legs and a broad carapace. They are the finest bur- rowers, and have, in addition to the well-developed cheliceral rake, rows of short digging spines on the first legs, which aid in scraping and cutting the soil. Their bodies are rounded and fit the burrow snugly, with the legs pressing closely against the sides. Their struc- ture bespeaks strength and ruggedness. As is true of most spiders, n4 AMERICAN SPIDERS they are active during the evening and at night, but they rarely leave their burrows. At the exit of its tube, holding the door ajar, sits the spider, ever watchful for the approach of food. On occa- sion it will rush forth to capture an insect, but most of its prey is taken without completely leaving the burrow. Inside the nest it is an agile creature; outside, a clumsy one. When disturbed, it closes the door firmly and holds the lid with chelicerae and claws, bracing its legs against the sides of the silken burrow. In this posi- tion, considerable force is necessary to dislodge it. Even with the aid of a knife blade one has difficulty in forcing the door. Two well-known genera, Actinoxia of the western and Myr- mekiaphila of the southeastern United States, make the type of nest Moggridge called "the double-door branched nest." The trap door of this nest is of the wafer variety. It is a thin, suborbicular cover almost wholly made up of silk, without layers of earth, and lies on the entrance rather than fitting into the aperture. It is not substantial enough to serve as an impregnable barrier to an intruder, being soft and pliable, and not heavy enough to fall over the open- ing if it is pushed back very far. It is only a superficial, hinged cover, which is camouflaged outside with moss, earth, or debris. The burrow proper is a cylinder lined with silk; but the particular innovation in this nest is the second burrow, a secret side chamber cleverly concealed by a trap door so constructed that it can close either the main tunnel or the side branch. (See Text Fig. 3, B.) The burrows of Alyrmekiaphila torreya are found on the leaf- mold-covered slopes of Torreya Ravine in Liberty County, Florida. This species digs a burrow that averages about ten inches deep. The nests are usually found in sandy soil penetrated by a maze of roots, and almost always contain at least one or more abrupt bends. Halfway down the tubes are the side chambers, one to a burrow, marked by wafer-type doors. The entrances to the outer burrows are lined with silk, and provided with a peculiar type of door, which, when standing open, is more like a silken collar than a trap door, but which takes on the appearance of a well-camouflaged trap door of the wafer type when closed by a slight push. This door the spiders sometimes leave standing open during both night and day. Most of the tarantulas that make an inner door are about two thirds of an inch in length. Their bodies (Plate XIV, male) are slimmer than that of Pachylomerus, ordinarily yellowish brown, and sparsely clothed with brown hairs. Their legs are longer and they PLATE XIII Female purse web spider, Atypus bicoLor Martin H. Muma PLATE XIV Lee Passinore A WESTERN TRAP-DOOR SPIDER, Aptostichus. DORSAL VIEW OF MALE '^ -:t :>£'>-^<-f''<*- George M. Bradt George M. Bradt a. Surprised in its burrow b. Exposed burrow A MEXICAN TRAP-DOOR SPIDER, Eucteniza TEXT FIG. 3. -TRAP DOOR SPIDER BURROWS (DIAGRAMMATIC) A. Cork door nest of Pachylomerus. B. Double-door of Myrmekiaphila. C. Nest of Cyclocosmict with spider in narrow recess. THE TARANTULAS 115 lack the rows of spines that the other group has, possessing instead a light-to-heavy scopula of hairs on the distal segments. The open door of torreya appeals to us as being virtually an invitation to enter. Atkinson, who studied a similar species in North Carolina, thought that the principal chamber was intended as a prison for ants that wandered in and were captured after closing the inner door. He called the genus Myrmekiaphila because these spiders build their nests near or even in anthills, and he believed that the ants make up a large part of the tarantula's food supply. Moggridge interpreted similar nests in terms of a protective device. A wasp, intent on paralyzing the spider and placing an egg on its body, finds a trap door, which may be open or which she may open or force, and enters in search of the spider. The spider mean- while rushes to the bottom of the burrow and closes the main tube with the inner trap door. Should the wasp persist, the spider crawls into the side chamber, moving the dual-purpose door to protect that opening. Once the main tube has been fully explored and found empty, the wasp may leave without discovering the inner chamber. Cyclocoswia truncata is a trap-door spider remarkable for the peculiar shape of its abdomen, and interesting in that it had been considered by many as the rarest spider in North America. It is a large, fat creature, rather closely related to Pachylomerus except for the abdominal structure. This round, leathery, caudally trun- cated organ, in the absence of actual observations, had led to in- triguing conjectures as to what use it is put by the spider. The initial description of truncata was made by Nicholas Mar- cellus Hentz, the father of American araneology, who in 1841 gave it the name of My gale truncata. His specimens, all of which were females and all since lost, came from Alabama. In his words: ". . . this spider dwells, like other species of this subgenus, in cylin- drical cavities in the earth. Though many specimens were found, I never saw the lid described by authors as closing the aperture of its dwelling. The very singular formation of its abdomen, which is as hard as leather behind, and which forms a perfect circle, in- duces me to believe that it closes with that part, its dwelling instead of with a lid, when in danger." What Hentz meant by "the lid described by authors" is inexplicable, unless he was referring to the lids of nests of closely related spiders, since, to our knowledge, he was the first man to see and record the species. Along with draw- ings of the animal, Hentz included a sketch of "the hole in which n6 AMERICAN SPIDERS it resides," a simple, circular aperture in the ground, unadorned by semblance of lid, turret, or silken structure of any kind. Did Hentz actually see the entrance to a burrow? Did he draw upon nature or his imagination as a model for this sketch? We know that he never saw a lid, and we can only surmise as to whether or not he saw the entrance. We next hear of truncata in 1871, when Ausserer created two new genera, Chorizops and Cyclocosmia, for spiders distinguished from their nearest relatives by possession of a truncated abdomen. My gale truncata was made the genotype of Cyclocosima. Later, a second species of the genus was discovered near Tonkin, Indo-China. Thus, Cyclocosmia truncata enjoys the distinction, along with the American alligator and other animals, of having its nearest relative in Asia. In his monumental work, American Spiders and Their Spinning- work, McCook treated the natural history of spiders in great detail. His chapter "Enemies and their influence on habit" speculates fur- ther on Cyclocosmia. Led on by the singular "adaptation" of the abdomen, and encouraged by the work of Hentz and Ausserer, McCook sees in this hard disk "one of the most curious examples of relation of structure to enemies, or perhaps of the reaction of hostile environment and agents upon structure." Relying solely upon Hentz for his information, but cautiously warning that Hentz's conjectures need confirmation, he agrees that it is not im- probable that truncata uses its abdomen as a door. He further appends a beautiful sketch of the spider in this imagined position, and remarks: ". . . and one may imagine the intellectual confusion of a pursuing enemy, which finds its prey suddenly disappearing within a hole in the ground, but which, when investigated, presents nothing but a level surface where certainly a hole ought to have been." Credit for the rediscovery of Cyclocosmia largely belongs to Dr. H. K. Wallace of the University of Florida, who found the well-hidden burrows in the bottom of Torreya Ravine. Other colo- nies discovered in Alabama and Tennessee have since widened the known distribution of these curious spiders. Cyclocosmia seems to prefer a rather steep slope in a shady, cool, somewhat damp loca- tion. The first burrows found were in a vertical bank protected by the overhanging roots of a large tree, a situation characteristic of the ravines in Torreya Park, where small streams have been actively eroding their courses. These exposed red and yellow, sandy THE TARANTULAS 117 clay surfaces are partially covered with mosses and liverworts. The burrows are straight, cylindrical, and almost vertical in every in- stance. They are enlarged for two-thirds their upper length, then narrow abrupty until they are exactly the diameter of the hard abdominal disk of the occupant. Specimens are usually found head- first in the bottoms of the burrows, presenting their armor plate to the intruder. In this position they fit the cylindrical cavity so nicely, and they hold on with their claws so tenaciously, that it is necessary to dig the earth away in order to extricate them without injury. When disturbed, some back up their burrows to where there is room for them to turn and present their fangs. The burrow of Cyclocosmia is covered by a hinged trap door, which is similar in shape to that of Pachylomerus but much thinner and quite flexible, thus belonging to the wafer type. Most of the doors appear to be located in and under leafmold on the sides of the banks, a circumstance that makes them difficult to locate. It has now been established that Cyclocosmia is simply another trap-door spider, but an extraordinary one; what, therefore, can we conclude regarding the previous interpretations by older stu- dents of the use of its abdomen, interpretations that have persisted even into recent books and papers? Obviously, it is disproved that the spider closes the top of the burrow with its abdomen. In addi- tion to the fact that there is a wafer trap door covering the en- trance, it is impossible for the abdomen to plug the outer opening, because of the difference in diameter. Cyclocosmia seemingly has two lines of defense against enemies: its well-hidden surface door and its ability to run down to the bottom of its burrow and com- pletely plug the tube. (See Text Fig. 3, C.) The protective devices of the trap-door spiders herein considered may be briefly reviewed as follows: Pachylomerus and Bothriocyr- tum rely upon a fortress guarded by a heavy cork door, which they hold shut with surprising strength. Myrmekiaphila and Actinoxia build a weak, flexible cover that serves only to keep out rain, but is well camouflaged; they depend upon the deception of the con- cealed side chamber deep within their burrow. Cyclocosmia trun- cata carefully hides the wafer door to its nest, and to intruders presents its tough body armor as a shield. Sheet-Web Tarantulas. The spiders of the family Dipluridae have followed a course in their development quite different from either the trap-door spiders or the tarantulas. They spin a silken n8 AMERICAN SPIDERS funnel in a crevice, under rocks, or in thick vegetable growth, and then continue the silk out over the ground as an expansive sheet. The spider hides in the funnel, and waits for insects to fall upon the funnel or become entangled in the sheet webbing, whereupon it rushes out and captures its prey. This type of web is called a sheet web; it is the same in general plan as those spun by the American grass spiders, by some wolf spiders, and by one group of atypical tarantulas. The spiders that use this device for capturing insects are usually agile creatures, which can rush to the location of their prey with great speed. Their movement on the flat sheet has, in a nice comparison, been likened to a skier gliding over the top of the snow, whereas the bulky insects make headway on the yielding silk like a man walking through heavy drifts. The sheet-web tarantulas are specialized creatures. They have developed the best eyesight of all the mygalomorph spiders. Their bodies are quite long and flat, and the tarsi of their long legs are provided with an unpaired claw, as in the trap-door spiders. Since much of their prey drops on the webs during the day, they hunt equally well then as at night. Their spinnerets are frequently greatly elongated and widely separated. The terminal segments of their long lateral spinnerets are provided with many small spools, from which can be spun a wide sheet of silk when the organs are moved from side to side. Except for two genera (Hexathele of New Zealand and Scotinoecus of Chile, which have six) only four spinnerets are present. Most of the diplurids live in the tropics, where large species and great sheet webs are conspicuous objects. In the United States occur only two genera, and they are not notable for size or for their web building. Microhexura is of particular interest because it is one of the smallest of all tarantulas, averaging about one-eighth inch in length. These tiny creatures carry their egg sacs around with them in their jaws, held beneath the body between the front legs much as in the fisher spiders. Three different species are now known, one from the high mountains of North Carolina and Ten- nessee and the other two recently discovered in the mountains of Washington and Idaho. Although we can probably assume that these tiny diplurids spin a sheet web, its exact character has not been observed. They live under pieces of bark, decaying wood, logs, and deep debris, in moist deciduous woods or fairly dense coniferous forests. The species of Evagrus are considerably larger than Microhex- THE TARANTULAS 119 ura, running half an inch, and even longer in the tropics. All the North American species build thin webs on the ground, especially in rocky situations, with the funnels hidden away in crevices. They are all pale yellow or light brown, excepting those from Mexico, which are mostly black. The male is remarkable in having the tibia of its second legs much swollen and armed near the middle with a heavy spur, which aids in holding the female during pairing. Tarantulas. Largest of all spiders are the immense hairy creatures of the family Theraphosidae, which Americans call "tarantulas." Although these mygalomorphs have nothing in common with the wolf spider of southern Europe, which truly deserves the name "tarantula," they have so completely usurped this appellation that an attempt to change it would be futile. In most of Spanish Amer- ica, the covering of hairs on the legs and bodies of these creatures has earned them the name of aranas peludas— "hairy spiders." Not inappropriately, they are dubbed by the Brazilians carangueigeiras, because of the long bony legs— especially of the males (Plate XV)— and their stance and gait give them a superficial resemblance to crabs. In Mexico, native Indian names have largely been displaced by "tarantula," which is applied to almost any large spider. But in Central America, where these creatures are reputed to be danger- ous to horses, they are still called aranas de caballo, or matacab olios. Outside the Americas, the tarantulas are widely referred to as "mygales," or "bird spiders." This latter name is inappropriate and largely inaccurate, because most of the species are ground loving and have little opportunity to attack birds in trees. No matter by what name the tarantulas are known, they excite the imagination because of their great size and notoriety. In the steaming jungles of northern South America live the largest and bulkiest representatives of the whole tribe, enormous creatures that have no peers for size anywhere else in the world. A male Thera- phosa from Montagne la Gabrielle, French Guiana, measured three inches from the front edge of the chelicerae to the end of the abdo- men, and had a leg span when fully extended of ten inches. This specimen, which was black all over and only moderately hairy, weighed nearly two ounces. An enormous female Lasiodora, from Manaos, Brazil, the bulkiest tarantula I have ever seen, had a body three and one-half inches long, and measured nine and one-half inches with the legs extended. Quite handsome in her clothing of fine brown hairs, she weighed almost three ounces. izo AMERICAN SPIDERS Our United States species are pygmies by comparison. A full- grown male of Aphonopelma from Arkansas was found by W. J. Baerg to weigh a little less than one-half ounce. The greatest total length of the carapace and abdomen of this specimen was about two inches. A representative female of the same species closely approxi- mated the male in weight and body length. Large females often weigh as much as two-thirds of an ounce after they have been well fed. The long legs of our southwestern males span six or even seven inches. Owing to their formidable appearance, the tarantulas have ac- quired the reputation of being dangerous. This reputation they do not live up to either in belligerence or in the virulence of their bite. For the most part, they are sluggish creatures, which attack only when goaded to an extreme. Although our species are credited in many accounts with being great jumpers, leaping is not their specialty, and they ordinarily strike over a distance of only a few inches. In point of fact, they make fine pets, and some quickly become so tame that they can be picked up and handled with ease. The venom of most seems to have little harmful effect on man, but the powerful chelicerae of large species are capable of producing painful wounds. About thirty species of tarantulas live within the limits of the United States, for the most part in the arid Southwest. Their ab- sence from Florida and the southeastern states is rather surprising, since that area is seemingly ideal for these hairy spiders. Their eastern limit is the Mississippi River, and they occur north to a line starting between Missouri and Arkansas and ending on the Pacific Coast in the San Francisco region. Tarantulas abound in the tropics and there have developed many interesting types. A few of them have become arboreal and move over the surface of trees with great facility, frequently nesting in bromeliads and other stations far above the ground. Even the ground-loving species are good climbers, since their tarsi are pro- vided with thick brushes of hairs, which enable them to climb a vertical pane of glass with ease. The tarantulas of our American Southwest (Plates 12 and 13; Plates XV, XVI and XVII) on the other hand, are more restricted in habit. They are all ground loving, and dig their own burrows or live in those abandoned by rodents. Once they have become attached to a burrow and its particular surroundings, they stay there during their whole life. The area in which they hunt is small, usually only a few feet on each side, and PLATE XV \ V .** a. Clambering over stone b. Portrait MALE TARANTULA, Aphonopelma Richard L. Cassell George M. Bradt Lee Passmore a. Female on desert soil b. Web-covered entrance to burrow . Female and egg sac in exposed burrow TARANTULA, Aphonopelma Passmore THE TARANTULAS 121 they rush back into the safety of their tunnel at the slightest dis- turbance. Rarely do they live in regions of dense forest or heavy undergrowth, preferring open areas on hillsides, mixed desert growth, or the fringe of cultivated lands. The burrow usually has a loose webbing at the entrance (Plate XVI), spun there after the night's hunt and indicating that the spider is at home. During the winter months the opening may be plugged with silk, leaves, and soil, and, in some instances, a little mound of earth surmounts it. All spiders need water, and tarantulas are no exception. Indeed, Baerg attributes the complete disapperance of a large colony of Mexican tarantulas near Tlahualilo, Durango, to a drop in the normal rainfall from nine to three inches. On the other hand, small quan- tities of water poured into the burrow will often bring the spider rushing out into the open— a procedure that affords an easy means of collecting them. The tarantulas in the damp rain forests of the American tropics frequently live above the ground, and after heavy rains may be seen wandering around in the open. Aversion for water may well have inspired some of these creatures to become arboreal, and thus escape regular deluges that they might have ex- perienced on the ground or below the surface. Tarantula burrows (Plate XVI) are often tunnels under large stones. Within spacious confines the mother spider spins a tremen- dous sheet, upon which she deposits her large eggs. She then covers them over with a second silken sheet and binds the edges together to form a flabby bag. For six or seven weeks she watches over this sac, occasionally bringing it to the entrance of the burrow to warm it in the direct sunlight, until finally the babies emerge. The spider- lings are gregarious, and they often remain in the burrow for some time after emergence; eventually they disperse by walking out of the hole and moving in all directions. Since they are much too large to balloon away on silken lines, they settle down in the general neighborhood of the burrow, hiding under chips and stones for a time and then occupying tiny burrows in the ground. As with all spiders, there is a tremendous mortality in the young stages: from each sac perhaps only a pair of tarantulas reach maturity. Adulthood for the spiderlings is very far in the future, since ten years are usually required for either sex to become sexually mature. The females and the immature males live in similar burrows in the ground, remaining virtually indistinguishable until the last molt, at which time some are surprisingly revealed as males. Many a large spider of this group has been kept in a cage for years, known by 122 AMERICAN SPIDERS some common feminine name, when suddenly its true sex becomes manifest. The males are much darker than the brownish females, often nearly black, and have an abdomen set with rusty red hairs. Their final transformation gives the males an entirely different outlook on life. Whereas they have been content for years to live in a dark burrow, they now desert it and wander over the country- side in search of mates. This activity occurs late in the year, from July into November, and during this period they may be seen cross- ing the highways of the Southwest, frequently in considerable numbers. Most of the tarantulas observed wandering in the open are males, and these are seen only during mating season. Few survive the year in which they become mature; many die a natural death, others are killed by the female during courtship or after mating. It is quite different with the females, whose unusual longevity has been pre- viously noted. Living to a ripe old age is quite an accomplishment, for taran- tulas are plagued by many enemies. Various rodents dig into their burrows, and, unmindful of the poisonous hairs, use the spiders for food. The young are preyed upon by many birds, and lizards, frogs, and toads, and some snakes find them quite suitable dietetically. Insidious enemies are the small-headed flies of the family Acroceri- dae known to confine their attentions exclusively to spiders, in the bodies of which they develop as voracious maggots. The species of Pepsis, giant metallic blue or greenish digger wasps with rusty wings, specialize in tarantulas, in fact occur only where these large spiders are found. Preferred prey because of their greater bulk, the females offer a far more generous supply of nutritional food to this predator than do the males. The long legs of the male seem to give him some degree of safety, and when he elevates his body high on his legs, the "tarantula hawk" has such difficulty in stinging him that she may abandon her efforts. On those occasions when the female tarantula ventures forth during the day, she is fair game for the great tarantula hawks (Plate 12 and Plate XVII). The details of the ensuing struggle, quite as unequal as in the case of the trap-door spiders, are given by Petrunkevitch: The Pepsis comes deliberately to the tarantula on the side of the cage and drives her down to the ground. The next mo- ment she closes in on her victim in the manner already de- THE TARANTULAS 123 scribed, and bending her abdomen under the venter of the tarantula, introduces the sting between the third and fourth right coxae, close to the sternum. The tarantula struggles vio- lently and rolls with the Pepsis over and over on the ground. After a few struggles, the Pepsis lets go her hold on the taran- tula, walks off a couple of paces, turns and comes directly toward the jaws of the tarantula. Without the slightest hesi- tation, she slips under the tarantula, which raises as high as she can on all her legs. The Pepsis grabs the fourth left leg with her mandibles. The tarantula tries to bite here enemy, but the Pepsis holds her off by pressing her feet against the feet of the spider, while at the same time continuing her hold on the fourth leg with her mandibles. Meanwhile, she bends her abdomen and searches for the place to pierce with her sting. Now she finds it. It is the same place as in the first specimen, that is, the articula- tion membrane between maxilla, first leg, sternum and lip. In a few seconds the tarantula is paralyzed. The position of the two is very remarkable. The tarantula sits in her normal way, but the Pepsis lies on her right side, head toward the posterior end of the tarantula, sting in the place mentioned. After at least half a minute, the Pepsis withdraws her sting and walks off. The tarantula remains motionless. Presently one leg of the tarantula moves. The Pepsis returns, climbs on the tarantula, in- serts her sting between the sternum and the third coxa and holds it there for about a minute.20 All that remains is the transport of the heavy spider, often weighing eight or ten times as much as the wasp, to its grave, which may have already been dug. Once the victim is within the pre- pared cavity, an egg is deposited on its abdomen and the burrow sealed up. The paralyzed spider provides a fresh food supply for the larva of the wasp, and, though remaining alive for months, will almost never recover from the effects of the venom. The tarantula reacts to its enemies in various ways. By throw- ing itself back on its haunches and elevating its head to expose for- midable fangs, it assumes a defensive attitude that may frighten away timid adversaries. If a tormenter persists in goading the spider, it often elevates its abdomen, and, working its hind legs rapidly, scrapes loose a small cloud of extremely fine abdominal 20 A. Petrunkevitch, "Tarantula versus Tarantula-Hawk: A Study in In- stinct," Journ. Exper. Zoo/., Vol. 45 (1926), p. 381. i24 AMERICAN SPIDERS hairs. When these come in contact with mucous membranes of the eyes or nose of mammals or man, a very disagreeable urtica- tion results, which persists for some time. In discouraging some types of enemies, such as small mammals, this may be effective, al- lowing the spider to escape while the aggressor is recovering from the effects of the poison and is still partially blinded. (The bald spot on the abdomen of tarantulas is often a result of a full use of this covering of poisonous hairs; after each molt the spider is provided with another even covering of hairs and setae.) Unfor- tunately, this protective device can have no effect on those insect enemies, the solitary wasps, which are most important as predators. The body hairs of tarantulas have long been known to have in urticating effect on the skin of man; in allergic individuals they often produce distressing symptoms. It is quite probable that a toxic substance is present on the hairs, and the effect is not en- tirely mechanical. Support for this view is seen in the fact that alcohol in which these spiders have been preserved is capable of producing the characteristic itching and stinging. Because all United States species are ground forms, their food consists largely of the animals available in their restricted hunting areas. Beetles and grasshoppers are most frequently captured, but many other kinds of insects, and such crawling creatures as sow bugs, some millipedes, and other spiders, fall to their lot. It is well known that our species will kill and eat frogs, toads, mice, and liz- ards in captivity, and it is reported that occasionally these small creatures are captured in natural surroundings. During the summer months the tarantula catches and eats insects almost every night, frequently gorging itself. On the other hand, long periods of fast- ing seem to have little effect on the spiders. In order to ascertain just how long they could go without food, Baerg kept several of them supplied only with water. One of the females lived two years and four months without food, and other females almost matched this record. Though the belief is more widely held than is justified, tarantulas have long been known to capture and feed on small birds. The first record of this behavior was published in 1705 by the Swiss natu- ralist Maria Sibylla Merian in her Metamorphosis Insectorian Suri- namensiitm. A fine color plate shows one of the South American my gales in the act of feeding on a hummingbird. The spider, a great brown creature said to belong to the genus Avicularia, has its fangs imbedded in the breast of the gaily colored bird, which has Walker Van Riper, Colorado Museum of Natural History Black widow, Latrodeclus mactans, with egg sac PLATE l8 Shamrock orb weaver, Aranea trifolium, on flower Joseph R. Swain THE TARANTULAS 125 been struck from its nest. Mme. Merian's report (which was re- ceived with considerable skepticism, since it was not believed at the time that any vertebrates could be consumed by spiders) was later followed by many claims that birds, lizards, and other animals were habitual prey of the great tarantulas and even of other smaller spi- ders. Corroboration of the early stories came in 1863 from H. W. Bates, in his book The Naturalist on the River Amazon. This tal- ented observer actually saw the capture and killing of one of two birds that were attacked, and very accurately depicted the spider in the act of feeding on it. Since that time, the debate has been con- cerned with the question of the capability of the spider actually to make use of the body of the vertebrate as food, not with its ability to capture it. That a powerful, predaceous creature, armed with strong fangs and potent venom, can kill a bird, a mammal, a snake, or a lizard is not an astonishing thing. The arboreal tarantula cannot differenti- ate between a bird or -a large insect, and makes its capture in exactly the same manner— by springing upon it and striking it with its fangs. Spiders predigest their food by flooding the wound with secretions from the maxillary and other glands, softening the tissue so it can be sucked into the body. The powerful buccal secretions are known to have a digestive effect on meat, so it is not strange that even the bodies of vertebrates can be taken through the small mouth open- ing. A tarantula can reduce the fat body and wings of a large satur- niid moth to an insignificant vestige, and do so thorough a job of it that one wonders if chitinous outer parts were not absorbed along with the softer portions. It can reduce the bulk of a fat mouse or the body of a small rattlesnake in the same way, feeding on the gruesome corpse for many hours. In the United States the lessened opportunity to capture small vertebrates has kept our tarantulas largely insect eaters— a quite different situation from that in Brazil, where the ground-loving species of Granrmostola and Lasiodora are believed to kill and feed on frogs, lizards, and small snakes in their natural surroundings. In captivity, these large spiders definitely preferred such small cold- blooded animals, and would generally pay no attention to various insects offered as food. While experimenting on spider venoms, Drs. Brazil and Vellard of Sao Paulo kept fifty of the tarantulas in good health for eighteen months on a diet of frogs, lizards, and snakes. Small rattlesnakes and the venomous Eothrops were killed and eaten as readily as any other kind of snake. 126 AMERICAN SPIDERS When a Grammostola and a young snake are put in a cage together, the spider tries to catch the snake by the head, and will hold on in spite of all efforts of the snake to shake it off. After a minute or two, the spider's poison begins to take effect and the snake becomes quiet. Beginning at the head, the spider crushes the snake with its mandibles and feeds upon the soft parts, sometimes taking twenty-four hours or more to suck the whole animal, leaving the remains in a shapeless mass.21 One of the interesting bits of folklore prevalent in Mexico and Central America is the legend of the matacaballo. For many years it has been a general belief that tarantulas bite the fetlocks of mules and horses and cause the loss of the hoof. According to the story, the spider hunts out the sleeping animal at night and takes a narrow strip of hair from above the hoof for its nest building, using an acid- like secretion to make the hair slough off more easily. The site of the injury then becomes inflamed, infection occurs, and the hoof is lost. In another version, all goes well unless the spider is disturbed and bites the hoof. In order to prevent hair clipping by the mata- caballoy the natives run their animals through a footbath of water covered with about an inch of crude oil. The tarantulas do not like oil-covered hair, so the animals gain temporary immunity from the presumed scourge. It is now known that this often fatal disease is actually caused by a bacillus that is very prevalent in the soils of Central America. During the rainy season, the skin of the hoof becomes chapped and the bacillus is able to enter through small abrasions. Needless to say, tarantulas use only their own white silk for their nests. THE ATYPICAL TARANTULAS One of the two principal branches of the My galomorphae has culminated in the Atypidae, the purse- web spiders; they are the namesakes of the series known as "the atypical tarantulas." This series includes the most generalized of all living spiders, the liphi- stiids (family Liphistiidae) , which have changed little since the late Paleozoic and are the last remnant of an ancient group that 21 J. H. Emerton, Psyche, 1925, Vol. 39, p. 60. (Part of English abstract of part of article by Vital Brazil and J. Vellard, Memorias do Institute do Butantan, 1925, Vol. II, and 1926, Vol. III.) THE TARANTULAS 127 failed to alter its form to cope with altered environment. More advanced offshoots from this same primitive stock are the sheet- web atypical tarantulas (family Mecicobothrndae), the folding- door tarantulas and relatives (family Accatymidae), and the above-mentioned purse-web spiders. The atypical tarantulas have paralleled in their development the other principal branch of the suborder, the tarantulas and trap-door spiders, and have matched rather closely their handiwork in silk. The outstanding characteristic of this whole series is the clear- cut visual evidence of segmentation on the dorsum of the abdomen. In the past, the abdominal tergites of the Liphistiidae have been hailed as evidence, along with the spinnerets and some other fea- tures, to set the family apart by a very wide margin from all other spiders, and place it in a separate suborder. This early evaluation of the liphistiids has become so fixed in the minds of most spider students that they have denied that any other living spiders are segmented in the adult stages. One has only to look at the abdomens of Antrodiaetus, Hexura, or At y pus to see tergites that differ little or not at all from those of Liphistius. And a study of the other features of these genera demonstrates with little question that the relationship between the more generalized Liphistius and its modern cousins is a real one. The atypical tarantulas are of moderate size, few of them ex- ceeding an inch in length, and in general form and appearance they resemble the typical trap-door spiders. Most of them are accom- plished burrowers, but only the folding-door tarantulas and close relatives have the chelicerae fitted with a rake of coarse teeth for digging. The unpaired claw is present on the tarsi, but no claw tufts or tarsal brushes have been developed. The full complement of eight spinnerets is present in the liphistiids, and the pudgy lateral pairs bear some resemblance to those of ancient spiders. The other atypical tarantulas long ago lost the anterior median pair. The per- sistence of the anterior lateral pair is noteworthy, since it is present elsewhere among the mygalomorph spiders only in one or two primitive members of the family Dipluridae. The anterior lateral spinnerets are two-segmented and functional in Aliatypus, uniseg- mented and small in Atypoides and most other genera, and com- pletely missing in Antrodiaetus. In some respects the atypical tarantulas have outdistanced the typical tarantulas, even though the physical heritage of the former includes more generalized features. The male palpus is provided n8 AMERICAN SPIDERS with a conductor of the embolus— a shield for the protection of the delicate tube found in none of the typical tarantulas and apparently similar to that found in the true spiders. The epigyna of the females all agree in having four primary seminal pouches, whereas in almost all higher My galomorphae and true spiders there are only two. The atypical tarantulas are hardy creatures that live much far- ther north in the United States than any of the typical tarantulas. Some of the folding-door tarantulas are common in our Pacific Northwest, and extend even into British Columbia and Alberta. In Europe At y pus is found in England, and the same species occurs in Denmark, a location that would place it above the 50th parallel north. In the United States Atypus is uncommon in the north but has been taken in Massachusetts and Wisconsin, well above the 4oth parallel. Liphistiids. The liphistiids are the most primitive of all living spiders, still maintainng the appearance and probably the funda- mental structure of their ancient Paleozoic forebears. They occur only in the Orient, but are reviewed here for comparison with the other atypical tarantulas, which are predominantly American. Li- phistius lives in hilly districts in the Malay States and adjacent Sumatra, where five species occur, and in similar situations in Burma and northern Indo-China, each of which has a single species. The genus Heptathela comprises a single species from the southernmost Japanese island of Kyushu and the Luchu Islands, and one from Shantung, China. A series of tergites, all of which are conspicuous, hardened plates set with rows of erect setae, is a striking feature of the liphistiids. All twelve primary abdominal segments can be recog- nized by external tergites in Heptathela, whereas at least nine are distinct in Liphistius. The generalized condition of the abdomen is further seen in the median position of the spinnerets. The great space between them and the anal tubercle represents those reduced segments behind the sixth that in higher spiders are completely incorporated into the tubercle. Four pairs of spinnerets are pres- ent, but the median pairs are greatly reduced in size. The lateral spinnerets of Liphistius are short, thick, fingers, with a large basal segment and an apical portion that is divided transversely into many small rings, thus said to be multisegmented. In the other atypical tarantulas, the spinnerets have shifted much farther back, but in THE TARANTULAS 129 no case do they reach the anal tubercle. In the typical tarantulas and in all true spiders, the posterior segments are so much telescoped or obliterated that the spinnerets and anal tubercle lie close to- gether. In Heptathela the posterior median spinnerets are reduced in size and fused into a single tiny colulus; so these spiders are usu- ally said to have seven spinnerets. The internal features of the abdomen are also of particular in- terest in this family. Five pairs of ostia are found in the heart of Liphistius, the fifth pair belonging to the sixth somite, and this number has not as yet been found in any other spiders. Although Heptathela appears to have a more primitive external segmentation, its ostia are reduced to four pairs, showing that the external fea- tures have not kept pace with internal changes. The liphistiids differ from the atypical tarantulas in several other respects than those enumerated above. The sternum is very narrow and unmarked by sigilla. The eyes are well developed and seem to be specialized rather than primitive, since the lateral ones are enlarged and the anterior median very much reduced in size. The coxa of the pedipalp does not have a maxillary lobe even slightly developed; in this respect the liphistiids agree with the majority of the typical tarantulas. The liphistiids live in burrows lined with silk, the entrance to which is closed by a simple trap door of the wafer type sometimes fastened down by the spider with threads from the inside. A num- ber of lines of heavy twisted silk radiate from the lower lip of the opening, serving as signal lines to warn the waiting spider of the approach of insects. Sometimes the whole tube is set in the open against the side of a wall, instead of being at least partially embedded in the soil. The trap door and any exposed part of the tube is covered with sand, thus to some extent camouflaged against the natural background. Some of the liphistiids live in caves. The food of these spiders usually consists of ground insects of various kinds, but the cave-dwelling variety often subsist on a single species of grasshopper or cricket. These primitive spiders are clumsy animals, which do not put down dragline threads as do all other spiders. Ordinarily they as- sume a stance with the front three pairs of legs directed forward, an attitude suited to life in a narrow silken tube. They are said to be awkward movers, and when "placed on their backs on a flat surface cannot right themselves." i3o AMERICAN SPIDERS Sheet-Web Atypical Tarantulas. It is of particular interest that among the atypical tarantulas we should find a group that parallels very closely the sheet-web tarantulas of the family Dipluridae. The hind spinnerets of these spiders are greatly elongated (par- ticularly the terminal segment, which is flexible) and rather widely spaced; this is probably an adaptation for spinning the sheet web, and it illustrates how in widely unrelated creatures similar activities often lead to the production of similar morphological features. The resemblance between Hexura and the diplurids is an amazing one. We find it running over a silken sheet web as do its distant relatives. Were we not deterred by what appear to be more fundamental features, we would ordinarily place them close together, perhaps deriving one directly from the other. As in most atypical tarantulas, six spinnerets are present, and the one-jointed anterior lateral pair is much reduced in size. The distance between the spinnerets and the anal tubercle is not so great in Hexura as in Antrodiaetus, but this can be attributed to the mi- gration of the spinneretts back and to the side— a frequent occur- rence in the Dipluridae. The abdomen is provided at the base with a large brown tergite. The chelicerae entirely lack a rake or digging instrument such as is developed in the folding-door tarantulas. The palpus of the male has a well-developed conductor of the embolus, and the whole organ closely resembles that of the other atypical tarantulas. First found in the state of Washington, the typical species is Hexura picea, a dusky-brown spider about one-fourth inch long. It lives under leaves, trash, and pieces of wood or back on the ground in pine woods, there building a loose sheet web in which it stays and over which it runs. The male has long projecting chelicerae armed with prominent spines. A second species of Hex- ura, which differs chiefly in its paler coloration, has been reported from northern California. The genus Mecicobothrium, on which the family name Meci- cobothriidae is based, is represented by a single known species in Peru. Folding-Door Tarantulas and Their Kin. Except for Ac cat y ma of Japan, a little-known genus that may be the same as our better- known Antrodiaetus, the members of the family Accatymidae are exclusively American. Two species are known to come only from California: the turret spider Atypoides, and Aliatypus, which covers THE TARANTULAS 131 its burrow with a trap door. The remaining genus, Antrodiaetus, has numerous species; they are widely distributed in the southern states right across the country, and in the mountain states and the Pacific Northwest are the commonest mygalomorph spiders. An important feature of this group is the possession of a distinct rake on the chelicerae. For this reason they have long been placed among the true trap-door spiders of the family Ctenizidae, a group they resemble closely, but one that has taken an entirely different route in its development. In Antrodiaetus the anterior lateral spin- nerets have been lost, but in the other two genera the six spinnerets are all present, with the same arrangement as in At y pus. The pres- ence of two, three, or four well-marked tergites at the base of the abdomen in both sexes is invariable; these are strikingly large and distinct, set with rows of transverse setae as in the liphistiids. The turret spider, Atypoides riversi, lives in the foothills of the Coast Ranges of California, and is found in abundance along shaded streams and in thickets in the San Francisco Bay region. Its turrets are well-known objects. They are ordinarily open at the top, lack- ing completely a closing flap or trap door, but on occasion will be completely spun over and closed with silk and debris. The burrow is very long, usually inclined, and is lined completely and rather heavily with white silk. The aerial portion may be only a short chimney, but quite often there is a long tube, which, penetrating thick grass, moss, or debris, finally terminates in the expanded white lip of the turret. The spider takes whatever building materials are handy— leaves, small twigs, moss, bits of lichen, pine needles— and fastens them on the outside of the silken collar. Often most in- geniously constructed, the turret provides an excellent lookout for the spider, which sits in the entrance at dusk and catches the insects that come within its reach. The turret spiders are about half an inch long, with yellowish brown carapaces and darker brown or purplish abdomens. A re- markable feature of the male is the presence of a long, projecting process on each chelicera, which probably is concerned with mating since no similar spur exists in the female. The tiny anterior lateral spinnerets are composed of a single joint, and, judging from their reduced size and lack of spinning equipment, are rapidly being aborted. The median groove of the carapace is a linear impression. The dorsal tergites on the abdomen are well marked, three being represented clearly in each sex. The second exclusively Calif ornian genus is Aliatypus, which 132 AMERICAN SPIDERS comprises a single known species having about the same range as its congener, the turret spider, and sometimes found in the same col- onies. The burrow of Aliatypus calif ornicus is comparatively long, and either goes straight down into the compact soil or is provided with pronounced bends. The silken lining is quite thin, but thickens around the opening, which is covered with a trap door of the wafer type. The burrows are usually found along roadside banks and streams, where the spider seems to prefer exposed soil only thinly covered with vegetation. The female Aliatypus resembles the turret spider, but has a some- what broader carapace marked by a round median groove. The male resembles the female quite closely, and completely lacks a distinctive spur on the chelicerae such as is present in Atypoides. The male palpi are thin appendages fully as long as the first pair of legs. In this genus the most interesting characteristic is that the anterior lateral spinnerets are nearly equal in size to the posterior median, and are also two-jointed. Well-developed spigots show that they are still functional appendages— a fact that marks them as the most generalized of all mygalomorph spinnerets, except those of the Liphistiidae. Since they are bisegmented, we can state with confidence that they are definitely the anterior lateral pair, and thus corroborate on direct evidence what has been the presumption of the majority of araneologists. The type genus of the family Antrodiaetus is in many respects the most highly developed. The short anterior lateral spinnerets, greatly reduced in size in Atypoides, have here been completely lost. The carapace has the median groove present as a longitudinal impression. In the males the abdominal dorsum has three distinct tergites above the base, and in the females one or more is present. The chelicerae of the males are armed with a prominent tubercle set with black setae. These spiders live in burrows, which may descend a foot or more in the soil, and which often have prominent bends. The upper part of the burrow is usually well lined with silk; in western species the opening is often concealed under stones or hidden in debris. As a result of their secretive habits and their well-hidden burrows, relatively few females are known in collections, whereas the males, which rove around in the late summer, are quite common. An ex- ception may be made for some .eastern species that dig their nests right in the open, and are easy to find. About a dozen species of Antrodiaetus have been described from THE TARANTULAS 133 various parts of the United States. Several are known from the Southeast, and one of these occurs rather commonly near Washing- ton, D. C In 1886 George F. Atkinson studied a species in North Carolina and, because of the singular means by which it closed its burrow, called it a "folding-door tarantula." There are two equal doors, each forming a half circle, which hang on semicircular hinges; when closed, they meet in a straight line over the middle of the hole. Each night the spider throws open its burrow, and each morning closes the doors, as shown in Plate 16. On the method of capturing its prey, Atkinson had the following to say: One evening I placed several ants in the jar containing the nest. When an ant approached, so near the door as to send a communication to the spider of its presence, the spider sprang to the entrance, caught a door with the anterior legs on either side, and pulled them nearly together, so that there was just space enough left for it to see the ant when it crossed the open- ing. When this happened, the spider threw the doors wide open, caught the ant, and in the twinkling of an eye had dropped back to the bottom of the tube with its game. This I saw repeated several times during the months of January and February.22 Purse-Web Spiders. In the low hammocks of Georgia and Florida lives one of the most remarkable members of the tarantula fauna. It has received the common name of "purse-web" spider from the resemblance its web bears to the silken purses so much favored by ladies over a century ago. In 1792 John Abbot, eminent entomologist and artist of Savannah, Georgia, first described the tubes of the species that bears his name: "This singular species makes a web like a money purse to the roots of large trees in the hammocks or swamps, five or six inches out of the ground, fastened to the tree, the other end in the ground about the same depth or deeper. To the bottom of that part in the ground the spider retreats. I imagine they come out and seek their food by night as I never observed one out of its web. In November their young ones in vast numbers cover the abdomen of the female and the abdomen then appears very shrunk. The male is the smallest, but has the longest nippers. Taken in March and is not common." Atypus abboti digs a deep burrow in the soil at the foot of a 22 G. F. Atkinson, "Descriptions of Some New Trap-Door Spiders, Their Notes and Food Habits," Entomologica Americana, Vol. 2 (1886), p. 116. 134 AMERICAN SPIDERS tree. This it lines with silk, then prolongs the silken lining up the side of the tree. The aerial tube (Plate 15) is securely fastened to the bark by threads, and in full-grown females is about ten inches long and three fourths of an inch wide. Smaller specimens spin cor- respondingly smaller tubes, which are almost invariably placed up- right against a tree. The top of the tube is open, but the silk is so flattened and pressed together that the natural opening seems to be closed. An even covering of sand and other fine material serves to color and darken the white silk and make it less conspicuous. In Florida the tubes are most often found attached to sweet gums, oaks, and magnolias in deep forest where the soil is damp and rich in organic material— although they have also been observed in dry woods where the sandy soil has little or no covering of humus. In Atypus bicolor, a large spider shown in Plate XIII, the tubes of old females are often eighteen inches long. This species occurs from Maryland south into western Florida, and westward into Mis- sissippi. They live for the most part in mesophytic woods. Near Quincy, Florida, I found them abundant in deep woods near a small stream. The tube of Atypus takes form in a characteristic manner. The female spins a small, horizontal funnel or cell on the surface of the soil, and from this base works both upward to lay out the aerial tube, and downward into the soil. The funnel is pierced above, and a two-inch section of vertical tube is set up against a tree. This design is accomplished by laying down many single lines and spin- ning the whole together into a strong fabric. The spider then begins excavating and spinning the subterranean part of her habita- tion. She molds the soil into small pellets, which she disposes of through the opening at the top of the aerial web. The covering of debris over the surface of the tube comes, surprisingly, from within the burrow— instead of being laid on from the outside: the sand and small particles are pressed outward through the web until the whole surface is evenly covered. After the first section of aerial tube is completed, another length is spun and coated with sand. Thus by sections the web moves up the side of the tree, until it attains the full length for the species. Like an iceberg, the finished tube penetrates the ground much farther than the length of its visible, aerial portion. It is heavily lined with silk, which becomes stronger day by day as the spinnerets constantly lay down their dense bands. The European species of Atypus have habits similar to our THE TARANTULAS 135 American forms, with this exception: they only rarely extend their nest up the side of a tree. Instead, they spin a very short aerial tube, about two inches long, which rests on the ground, is suspended in the grasses, or is attached to stones. The end of this tube is closed and, as in our species, the spider never leaves the web. The leath- ery tube, rendered less conspicuous by its covering of sand and debris, would seem to afford considerable protection to the spider; this seems to be borne out by the fact that the atypids are largely immune to the attacks of pompilid spider wasps. Because the areial purse web is completely enclosed, and continuous with its subter- ranean portion, predators must cut through the web to locate the spider. The purse- web spider remains just inside the subterranean por- tion of her nest while waiting for prey, but at the slightest notice of a passing insect she moves into the aerial web. Her course is charted by the movement of the tube, and when the insect crawls over the surface, she rushes to the proper point and strikes her long fangs through the web, around or into the body of her prey. Hold- ing it until completely subdued, she at the same time cuts the tube and pulls it inside. A slight rent is left in the silk, which will later be sewed together, and in due time covered over with sand so evenly that no sign of the break is evident. A tidy housekeeper, Atypus when through feeding brings the shrunken remnant of her prey to the opening at the top of her web and casts it out. In the same way, she voids her milky white, liquid fecal material through the opening— with such force that it is shot several inches away. In June the males become adult and leave their webs to wander in search of a mate. Until the time they become fully adult they live in nests that are to all appearances identical with those of the females, and occasionally in season they can still be found in their tubes. The mating behavior of our American species has not been described, but it is probably similar to that of the better-known European types. When the male finds the tube of a female, he drums upon it with his palpi, and presumably is able to ascertain by the reactions of the female to this drumming, whether he is go- ing to be welcome. After a short period, he cuts open the tube and enters, and the break is repaired by the female. Mating occurs deep in the tube. It is believed that the male lives in the burrow for many months before he dies. The eggs are deposited within the burrow, and hatch during the summer months. The young may stay with the female for long periods, but in most instances they 136 AMERICAN SPIDERS leave the nest in the late summer and disperse by ballooning. Their threads are usually a heavy band of parallel strands less fine than those of true spiders. Some of the young, it is thought, do not take to the air but merely walk a short distance from the maternal nest and begin work on tiny tubes of their own. The purse-web spiders are the most extraordinary of all the atypical tarantulas, as regards both their physical features and their singular habits. The marks of their primitive origin are clearly shown in the presence, above the base of the abdomen in each sex, of a single well-marked tergite, and the considerable separation of the anal tubercle from the posterior spinnerets. In the reduction of their cardiac ostia to three pairs is clear evidence that their heart has become specialized, or simplified, at a much faster rate than have other features of the abdomen. The chelicerae, though not provided with a rake for digging, are modified into effective shov- els for carrying loads of sand or pellets of soil. The fang is a long, thin spine well designed to pierce the silk and hold the prey. The species of Atypus are found in the north temperate zones of Europe, Japan, and the eastern United States. Species are also known from Java and Burma in the eastern tropics. Another genus of the same family is Calommata, which is largely restricted to tropical areas in Africa and the Orient. The four American species of Atypus are all confined to the eastern portion of the United States, and are most abundant in the extreme southeastern part of their range. Only three of our species are well known, and only two of these moderately common. The females of all the American species are predominantly brown in color, shining, and only very sparsely set with covering hairs. The robust body is provided with quite short legs and long chelic- erae, and runs about half an inch in length— although bicolor, the largest known species is often an inch in length. The males are similar to the females in most respects, but have longer legs. In niger, a shining black spider most closely related to the European species, the disparity in size of the sexes is not particularly marked; but in the other species the males are somewhat smaller than their females, and very brightly colored. The abdomen of Atypus abboti is a beautiful iridescent blue or purple, set against black legs and carapace. Atypus bicolor has carmine legs, which, contrasting with its deep-black carapace and abdomen, make it the most striking of all our species. CHAPTER VIII The Cribellate Spiders T J. HE TRUE SPIDERS THAT POSSESS a flat spinning organ close in front of the anterior spinnerets are called "cribellate spiders." This organ, which exists in addition to the usual six spinnerets, is known as the "cribellum." It is the homologue of the anterior median spinnerets, and has been retained as a functional spinning organ, whereas in other true spiders it is represented by an inconspicuous vestige. The cribellum (Text Fig. 4, C) may be likened to the fused spin- ning fields of two spinnerets lying nearly flat against the surface of the abdomen, all but the tips of the originally paired fingers having disappeared. The dual character of the organ usually is evident on close examination, which shows an actual division of the field by a longitudinal line or ridge, or a pinching at the point of division. The spinning field itself is covered by thousands of tiny spinning open- ings, which give it a sievelike appearance under magnification, and from which come exceedingly fine threads of viscid silk. The ordi- nary silken threads of cribellate spiders are derived from glands opening on spinnerets, as in other spiders. Whenever cribellar silk is combined with the regular threads, the line becomes so character- istic in color and physical appearance that it is called a "hackled band." Invariably accompanying the cribellum is an accessory comb of hairs called the "calamistrum." This is a line of curved setae, dif- fering somewhat in appearance in the various families, and always found upon the metatarsis of the hind legs (Text Fig. 4, E). The use of the cribellum and calamistrum together as a spinning and carding apparatus to produce the cribellate thread is essentially the same among all the spiders of the group. Let us consider the method of a typical hackled band weaver of the genus Amaurobius. The cribellum of Amaurobius is divided longitudinally by a 138 AMERICAN SPIDERS distinct septum, on each side of which lies a spinning field. Because of this division, the hackled band spun by Amaurobius consists of two ribbons instead of the one band usually found in the cribellates that have obliterated the limits between the two spinnerets. The two ribbons are borne by two strands of dry silk presumed to come from the ampullate glands. To spin its composite hackled band, Amaurobius holds the hind leg of one side at right angles to the long axis of its body, with the tarsus resting against the metatarsus of the leg of the other side. The metatarsal comb is then rubbed back and forth over the cribellum, drawing out two ribbons that are attached to two lines of dry silk coming at the same time from the spin- nerets. After a period of incessant spinning, the spider shifts to the other leg, supporting it as before by resting the tip across the oppos- ing metatarsus. The result is a fairly regular, ribboned band of silk that seems to the naked eye a single thread, and has a characteristic bluish color. The cribellate spiders have retained more units of spinning equipment than have any other true spiders, and have maintained all of them as functional organs. It is thus not surprising that none have become truly vagrant, and that all rely to a very large extent on their viscid threads to capture insects. The very fact that they have retained the cribellum, with its glands of sticky silk, indicates their reliance on it in some measure to entangle their prey. In the cribellates, the unpaired claw on the tarsus is usually pres- ent, but it is lacking or reduced in size in a few members that may be taking their first steps toward becoming vagrant types, or have learned to do without these tarsal hooks in their webs. Some cribel- lates are confirmed aerial spiders and spin tangled webs, sheet webs, and orb webs, from which they hang downward. Others run over an irregular blanket of webbing in an upright position. The cribel- late spiders have produced web structures closely paralleling those of the ecribellate spiders, the only difference being the use of the hackled band by the former. The cribellates are quite sociable creatures. During the mating season the males enter the webs of the females and live there as partners until— presumably— they die a natural death. This tolerance carries even beyond the mating season, for among the cribellate spiders we find nearly all our social spiders. Some members of al- most every family are known to live together, in colonies similar to those of certain gregarious caterpillars. One of the controversial and perplexing problems in spider TEXT FIG. 4.-THE CRIBELLATE SPIDERS A. Line trap of Miagrammopes. B. Triangle web of Hyptiotes. C. Cribel- lum and spinnerets of Amaurobius. D. Tarsal claws of Amaurobius. E. Metatarsus and tarsus of Hyptiotes, showing calamistrum. F. Retiarius snare of M enneus (after Ackerman). PLATE Richard L. Cassell Banded Argiope, Argiope trifasciata, with swathed prey, dorsal view PLATE 20 Richard L, Cassell Banded Argiope, Argiope trifasciata, ventral view THE CRIBELLATE SPIDERS 139 phylogeny has to do with the origin of the cribellate spiders and their relationship to the ecribellate families. By some they are held to be a homogeneous group derived from a single line of ancestral spiders that put their fading anterior median spinnerets to a new and original use by inventing the calamistrum. On the other hand, the cribellate spiders can be regarded as a remnant held over from a time when all spiders were cribellate, and the modern forms can then conceivably originate from one or several distinct lines. The presence among these spiders of some in which the second pair of book lungs is still persistent suggests a very early origin for the group, and also strongly indicates that all ancestral spiders were provided with cribellum and calamistrum. If we subscribe to this belief, then it can be put that the ecribellate spiders have lost the spinning organs, rather than that the cribellates have gained them. In this book we consider the cribellate spiders in a single chapter, even though among them are certain discordant elements that sug- gest a multiple origin. THE FOUR-LUNGED TRUE SPIDERS One of the oldest American spiders is Hypochilus thorelli, a strange relic of the past, whose forebears were probably aerial con- temporaries of the Paleozoic ground tarantulas. The only known living relatives of Hypochilus are two species of a related genus, Ectatosticta, found in China and Tasmania. Although we regard spiders of the family Hypochilidae as being true spiders, they share many of the features of the tarantulas, the most notable being pos- session of the posterior pair of book lungs, which in all other true spiders have been transformed into tracheae. These lungs are situated beneath and about at the middle of the abdomen, and their spiracles open at the sides of a prominent furrow. The chelicerae are pro- vided with venom glands entirely contained within the basal seg- ment, and the heart has four pairs of ostia as in the tarantulas. Perhaps the most distinctive badge of the true spider is the articula- tion of the chelicerae. In Hypochilus we find the chelicerae largely intermediate in type between those of the true spiders and those of the tarantulas; since the claws do not point toward each other, they are in many respects nearer those of the latter. The cribellum of this spider is a rounded plate lacking the median dividing ridge but pinched before and behind to indicate its original dual character. i4o AMERICAN SPIDERS It sits on a low elevation that strongly suggests the segment of an ordinary spinneret. Hypochilus shows a greater difference from the tarantulas in its habits of life than in its physical features. Whereas no tarantula has become a confirmed aerial cobweb spider, the hypochilids and a great many other true spiders have. It is quite possible that spiders resembling the hypochilids were the first to break away from the conservative tarantulas, and that Hypochilus thorelli is a modern representative of an ancient group that gave rise to all true spiders. Hypochilus has a very restricted range. It is found in the can- yons of the mountains of the southeastern United States, where it is quite abundant at elevations from one to about five thousand feet, and especially so in the Great Smoky, the Nantahala, and the south- ern half of the Blue Ridge Mountains. It prefers dark situations under overhanging rocks, and natural arches in forested areas near streams. Its webs— conspicuous objects even from a distance— are often found close together under the rock ledges. They are shaped like lampshades with the top pressed against the overhanging sur- face, and consist of a very heavy mesh of cribellate threads over a basis of dry silken lines. The spider hangs underneath this net with its long legs touching the sides of the aerial portion. It resembles most closely the true spiders of the family Pholcidae, and is remarkable for the great length of its banded legs. Hypochilus does not seem to have the power of autotomy, and its legs do not break off at a point weak- ened and predetermined for this purpose, as in other spiders. The males, which mature in the fall, differ little from their dull, yellowish mates. The male palpus is of very generalized design, and is pro- vided with a conductor of the embolus as in the atypical tarantulas and most true spiders. The epigynum of the female is quite simple and presents no external openings. In the internal epigynum are four receptacles— a feature shared by the atypical tarantulas and some of the most primitive true spiders. THE FILISTATIDS One of the very common house spiders of the southern Ameri- can states is Filistata hibernalis, a large animal whose webs are often prominently outlined on the outside walls of buildings. This spider hides in a crevice during the day, and at night comes out to spin THE CRIBELLATE SPIDERS 141 on its web, placed over the retreat as a rounded net, which soon gathers to its sticky lines an unkempt covering of dust and debris. The web, often more than a foot in diameter, is composed of a series of regular, radiating lines of dry silk over which has been spun many lines of cribellate bands. The touch of an insect vibrates the web, and the disturbance is communicated to the hiding spider. The hackled band of Fihstata is composed of four different kinds of silk. The cribellum is combed with a very short calamis- trum, and many tiny loops are produced, which, bundled together, give a most irregular shape to the characteristic threads. The spider is said to lay down a dry line of two threads, to retrace its steps upon this, and then to put down the irregular hackled lines, thus accomplishing its purpose in three operations rather than in a single one as does Amaurobius. The female of the common house Filistata is about half an inch long, and is quite variable in color, being light brown, dark brown, or often velvety black. Older specimens are usually much darker than the young ones. The male is pale yellow, smaller, and his slender body is fitted with much longer legs, which, during court- ship, he uses to hold the front legs of the female as the couple parades back and forth in a prenuptial dance. The members of this family are considered quite generalized spiders because of the simplicity of their palpi. Nearly a dozen species, some very much smaller than hibernalis, occur in the ex- treme southern parts of the United States. THE TYPICAL HACKLED BAND WEAVERS In this group are included the great majority of cribellate spiders and almost all of those that occur in the temperate regions. About two hundred species are known from North America, but for the most part these represent only a few different types. The preva- lence of their meshed webs on the ground and on vegetation every- where is an index of their abundance and comparative success. With few exceptions, they are confirmed web spiders and stay in their snares most of the time, walking upright over the bluish sheet. Their sedentary bent has not molded them into such aerial types as in the Uloboridae, but some do climb vertical meshes and are at home in aerial tangles. Three claws are almost invariably present on the tarsi, but they are never aided in their climbing by accessory i42 AMERICAN SPIDERS claws. Many are swift runners that can be seen dashing across paths with the celerity of vagrants. Their habits are thoroughly interest- ing, although they are not known to do things quite as amazing as do some of the aerial web spinners. Three different families occur in American fauna, the Dictyni- dae, Zoropsidae, and Oecobiidae; two other families of somewhat greater interest are not found within our territory. One of these latter, the Eresidae, is made up of robust, moderate to large crea- tures, similar to jumping spiders, and often quite brightly colored, especially the males. Some of them are fine tunnelers and are even known to use a trap door to close their tubes. Many spin sheet webs connected with tubular retreats. Some species of Stegodyphus join together and spin an immense communal web over bushes, forming an irregular saccular retreat partitioned in various ways, in which many individuals live amicably together. The second exotic family is the Psechridae, whose only repre- sentation in the New World is one small group from Mexico, whereas several large conspicuous types are common in the Pacific regions. These spiders will often spin a huge web, at the center of which is a flat sheet similar to those made by the sheet-weaving Linyphiidae. The spiders creep over the ventral surface of the sheet, hanging back-downward, as do the aerial ecribellate spiders, on greatly elongated legs of which the terminal segments are flexible. One of the strange features of the psechrids is the presence of well- developed claw tufts on the tarsi. These are probably used for a very different purpose than are the claw tufts of the wandering spiders. Perhaps they aid in unfastening the median claw, and serve in the same way as the accessory claws of the orb weavers. Among the largest typical cribellate weavers of the American Dictynidae are the ground spiders of the genus Amaurobius. Many of the females are robust creatures attaining a length of three-fourths of an inch. Their colors are usually brown or black, but the dorsum of the abdomen is variegated, with a series of yellowish chevrons forming a pale band. The males, not far inferior in size, are usually in evidence only during the fall and very early spring, and are rarely seen during the rest of the year. A single native species— Amaurobius bennetti—is common in the eastern part of the United States, but many others abound in the mountains of the western states and in the northern woods. Amaurobius spins a large irregular web in dark, moist situations. Whereas much of the silk may be hidden from sight, not infre- PLATE XVII . The tarantula assumes a defensive attitude . •7 &>' •••••• W^i £.«« Passmore Lee Pass more b. The wasp inserts its sting c. Pulling the bulky prey to prepared burrow TARANTULA, Aphonopelma, AND TARANTULA HAWK PLATE XVIII Lee Passmore a. Female and pygmy male b. Egg sac SILVER ARGIOPE, Argiope argentata Lee Passmore Lee Passmore Lee Passmore a. Banded Argiope, b. Humped orb weaver, Argiope trifasciata Aranea gemmoides EGG SAGS OF ORB WEAVERS THE CRIBELLATE SPIDERS 143 quently the web is placed in plain view against a vertical surface. The dry framework of the snare is commonly put down as a series of lines from the central retreat, in which the spider stays most of the time. Over the dry lines the cribellate silk is spun loosely, mak- ing a thick mat upon which the spider runs. The spinning activities are best observed at night; then the carding can easily be seen in the rays of a flashlight. At this same time males may be found near the female web. Amaurobius' egg sac is a flattened bag, attached to a stone and usually covered over with a mesh of threads. The females stay with the eggs for long periods, often being found with the sac some time after the young have hatched. One of the commonest members of the genus in the eastern United States is the domestic Amaurobius ferox, an immigrant from Europe. This spider, which is much darker and somewhat larger than the common bennetti, lives in cellars, under floors of houses, and under wood and debris near human habitations. It is rarely found far from man. Australia is particularly well supplied with spiders related to Amaurobius; the habits of certain varieties are of special interest. In the Jenolan Caves of New South Wales lives a gregarious species, Amaurobius socialis, which spins great webs on the roof. These giant reticles, one of which measured twenty feet in length and more than four feet at its greatest width, hang from the roof and are draped over the stalactites. They are closely and densely woven to the consistency of a heavy fabric, such as a shawl, and are filled with openings through which the spiders retreat to the interior. Mating, egg-laying, and emergence of the young all occur within the limits of the web, as among many truly gregarious spiders. Also in New South Wales may be found certain gregarious spiders, perhaps relatives of Amaurobius^ which infest orange groves. They mat the limbs with a thick covering of web so densely woven that it affects the normal respiration and development of the tree. The leaves wither and fall, the dead branches are left, covered with unsightly webbing. Such spiders can become a pest of nearly equal malevolence to tent-building insects. In the genus Dictyna are small spiders, averaging one eighth of an inch in length, which are known by the unwieldly name of "lesser mesh web spinners." Some few of these are brightly colored in reds, browns, and tans, with here and there a brilliant yellow spot, but for the most part their bodies are dull, clad plainly in gray 144 AMERICAN SPIDERS hairs. Though several different groups of these small spinners are known, they are all similar in appearance and in habits. Some of the smallest live under debris on the ground, where they spin tiny webs and are rarely noticed; others of larger size spin on the walls of buildings, and on plants. Their lacy meshes are conspicuous ob- jects, but, aging, become obscured with dust and lint. Since ap- proximately one hundred different species of Dictyna are known from North America, mention can be made here only of a few species whose habits are illustrative of the whole group. Dictyna annulipes (formerly muraria), a small species with a large oval abdomen, has its dull body quite completely masked by a covering of light gray hairs, which on the carapace form three distinct stripes, and on the abdomen outline a pattern of darker chevrons. Favorite sites for its web are board fences and the walls of buildings. A tiny crevice between boards will provide this spider with an adequate retreat from which it can lay out the dry founda- tion lines of its snare. These lines frequently radiate with the most precise regularity, and, when crossed evenly with the thick hackled bands, reveal a web as delicately spun as a lace doily. One observes that annulipes often chooses the outside of a window sash as a location for its snare, and lashes it to the smooth glass as well as to the adjacent wood. Early summer is usually the pairing season for these friendly little spiders, and at this season the male may been seen in the web of the female. He resembles her closely in general appearance, but is somewhat more slender and has longer legs. His head is aften quite elevated, arching over the long, curved chelicerae. These latter are provided with a stout spur near the base; they turn up- ward at the tip, and curve strongly outward at the middle, leaving a conspicuous opening between. In some species of this particular group, the conspicuous chelicerae are known to be used to hold the jaws of the female during the mating, and it may be presumed that they render the same service for our own species. A close relative of annulipes is Dictyna valuer ipes, a slightly larger spider, similarly clothed in pleasing gray raiment, which pre- fers open sunny fields for its home. The usual sites are the ends of weeds and grasses (Plate III), and especially the dried stems and stalks left over from the previous growing season. Upon this har- vest skeleton volucripes spins its characteristic mesh; the foundation lines bridge from stem to stem, and over them is woven a criss-cross of viscid bands, to form perfect little lattices and other pleasing THE CR1BELLATE SPIDERS 145 symmetries. During the summer white, lens-shaped egg sacs are hung in the deeper parts of the tangle, and after the young hatch they spend some time in the web with the mother. Much more brightly colored than either of the above-mentioned species is Dictyna sublata, often light brown in color with its oval abdomen marked in yellow above, and its legs almost white. Sub- lata hides its web in the leaves of bushes instead of placing it in the sun. It will find a leaf with slightly rolled edges, then spin a thin, sheetlike web across the opening to form a shallow bowl; in this it remains and here its egg sacs are placed. The villagers of certain mountainous portions of Michoacan, Mexico, are plagued during the rainy season by immense swarms of flies that invade their homes. Their defense against these pests is unique. They rely upon the mosquero (Coenothele gregalis), a tiny cribellate spider one sixth of an inch long, which lives in vast colonies on the twisted oaks and scrub trees at altitudes of about eight thousand feet. The nest of a mosquero community is often more than six feet square, and thickly invests each branch of an entire tree with a spongy inner layer of dry silken lines and an outer envelope of sticky hackled-band threads. The villagers cut a branch from the tree, and suspend the animated fly trap from the ceilings of their homes. The accommodating houseflies alight on the sticky threads, whereupon they are enveloped and dragged into the inner galleries to become the prey of the colony. After the fly season is over and the spiders have become mature, the adults desert the colonial web, perhaps to start new colonies elsewhere. Their eggs and young remain, develop in the inherited nest, and are on hand during the next fly season. In the field the webs of the mos- quero resemble those of processionary caterpillars. In the inner recesses of the communal web live many small beetles of the genus Melanopthalma said to attend to the cleanliness of the nest by keeping it free of debris. These commensals live on the small bits of food discarded by the mosquero. Also living in complete harmony with the colony is one of the running spiders, Poecilochroa convictrix, which is also supposed to be a commensal though its exact status is less certain. The presence of thick brushes of scopular hairs beneath the metatarsi and tarsi of the cribellates of the family Zoropsidae, and the absence or great reduction in size of the median claw, would seem to indicate that these spiders are hunters. Indeed, they are often compared to the hunting clubionids, which they resemble in 146 AMERICAN SPIDERS general appearance and in superficial features. However, Zoro- cratesy the only American representative, still relies on an expan- sive web to snare its prey. The net, resembling to some extent that of Amaurobiusy is usually placed beneath stones, and made into an effective trap by spinning many viscid hackled bands over its dry framework. It may be that Zorocrates' scopular brushes contribute to better movement over the surface of its web, or perhaps that this spider is on its way to becoming a vagrant form and therefore spends part of its time outside the limits of its snare. Several species of Zorocrates live in our southwestern states, but they are still little- known creatures. In the related group of cribellate spiders, Acanthoctenus, known only from tropical America, true tarsal claw tufts are present in addition to the thick scopular brushes beneath the metatarsi and tibiae. These flattened creatures often sit under bark, closely ap- pressed to the surface, and move with great speed when they are touched, their claw tufts aiding them in holding on to surfaces, as with the ecribellate vagrants. Acanthoctenus combines a sedentary aptitude with running ability. It spins a loose web, embellished with sticky bands to entangle its prey. One of the few cribellates that has attained a nearly cosmopoli- tan distribution is Oecobius annulipes, a tiny spider less than one eighth of an inch in length, which is one of the few American repre- sentatives of the curious family Oecobiidae. The generic name of this spider signifies "living at home," and well characterizes these dwarfs found in and on the walls of dwelling places. The micro- scopic webs of Oecobius are frequently spun over cracks in the sides of buildings; they are only about the size of a postage stamp, but seem quite adequate to entangle the tiny insects used for food. The spider, which is pale white or pale brown and marked with distinct black points, is common in the southern part of the United States and has long gone under the name parietalis, but it is now known to be identical with the universally distributed annulipes. Several related species of Oecobius ocur in the southern part of this country, living under stones, on trees, or on buildings. THE AERIAL HACKLED BAND WEAVERS The most pronouncedly aerial of all the cribellate spiders are those of the families Deinopidae and Uloboridae, groups largely THE CRIBELLATE SPIDERS 147 tropical in distribution that press northward in small numbers into the temperate zone. Within the limits of the United States are found representative species, many of which are remarkable for their physical appearance and strikingly resemble bits of dried leaves, twigs, thorns, buds, scales, and similar natural objects. The common name of "stick spider" has been applied to some of them; the name characterizes the whole group rather well, even though not all are elongate. All do hang downward from a more or less intricate web, and in their movement on silken lines parallel closely such aerial ecribellates as the orb weavers and tangled web weavers. By some students these spiders are regarded as closely allied to their ecribel- late cousins, but the extensive use of the hackled band sets them apart and indicates only a distant relationship. The third tarsal claw is present, modified to suit the needs of confirmed sedentary types. In this series of species have been developed some most ingenious devices for capturing prey. The ogre-faced spiders have perfected a method of expanding and hurling a sticky net over flying insects. Not to be outdone by their cousins, the uloborids construct a splen- did orb web rivaling in excellence that of the orb weavers of the family Argiopidae. The triangle spider, Hyptiotes, has abandoned all the orb save a single sector of four rays. Even more niggardly is the tropical stick spider, Miagrammopes, which employs but a single line on which it spins a band of sticky silk. Ogre-Faced Stick Spiders. The name "ogre-faced spider" is applied appropriately to species of Deinopis (from the Greek, mean- ing "terrible appearance") because of their weird aspect and the enormous size of their posterior median eyes, which, projecting forward like great headlights, render inconspicuous the remaining six. The habits of these spiders, particularly their nocturnal net casting, would seem to demand good night vision, and this doubt- less accounts for the development of such large eyes. The American species of Deinopis is quite rare, uncommon even in the extreme southeastern portion of the country and is apparently the only mem- ber of the family reported from the United States. Quite a number of species, many with grotesquely formed heads and humped and lobed abdomens, are known from the tropical region all around the world. Nothing has been published on the habits of our ogre-faced spider, but it seems certain that it will be found to perform in the same way as the species of Menneus and Deinopis in Africa and Australia. 148 AMERICAN SPIDERS This American ogre-faced spider, Deinopis spinosus, is a slender creature, which hangs from a small web of dry silk on very long, stiltlike legs during its casting operations. When mature, the female is about two thirds of an inch long, and frequently has as .a notable feature an abdomen armed above and near the middle with short projections. The male is smaller and more slender than his mate, and his thin legs are at least three times as long as his whole body. A dark band running the length of the abdomen below, and a few lines and spots above, are the only distinctive pattern on the other- wise drably marked bodies of these spiders. During daylight hours the ogre-faced spiders are usually to be found pressed flat against the bark of a branch near their snares. They assume a characteristic position: the forelegs are stretched out in front along a twig, while the hind legs grasp the twig and hold the body firmly. The resemblance of this spider to a bud, spine, or some other natural irregularity in the bark is a most striking one, and must certainly pay dividends by giving the creature some im- munity from predators. Completely quiescent during the day, Deinopis rouses to action at sundown, moves into the small tangle of dry silken lines, and prepares its capturing web. Conrad Ackerman has described in fine detail how Menneus, an African ogre-faced spider, spins her web and captures her prey. The animal lays down a horizontal foundation line, and from this stretches parallel vertical lines down and across to outline a rec- tangular base, all of dry silk. Across this base she spins a series of transverse bands of sticky silk, which she cards from her cribellum in the normal manner. The result of this latter operation is a small reticle of sticky lines about the dimensions of a postage stamp, which Mennens grasps in her four long front legs while with her hind legs she holds herself securely to the dry lines of the web. In this position, hanging back-downward, the spider waits for a night- flying insect— usually a moth— to arrive within the limits of her casting area. (See Text Fig. 4, F.) When her prey comes within reach, Menneus suddenly stretches the elastic snare to its full expansion, which appears to be five or six times its size when closed, and hurls herself for- ward, throwing the net over the moth and closing it down upon it with her four front legs. The moth is helpless and the spider at once bites it. After waiting a few moments, she carefully extracts it from the web and the insect does not move, probably THE CRIBELLATE SPIDERS 149 because of the paralyzing effect of the poison injected at the bite; but to make certain that it cannot escape, the moth is en- shrouded in silk spun across it, the hind legs drawing out the silk from the spinners and applying it to the insect. This method of snaring a victim was compared by Ackerman to enveloping it as the Retiarius with his net enveloped his oppo- nent before piercing with his trident in the Roman gladiatorial combats, or better, like the old-fashioned butterfly net on two sticks, held by the two hands, which was thrown over an insect to catch it.23 A single web suffices Menneus for a whole night of casting. After removing a victim from the sticky lines, she pulls and read- justs until the web again takes its essential rectangular shape. Then she resumes her vigil, feeding on the unlucky insect while she holds what may be only a tattered remnant of her snare. When the spider has satisfied her appetite, and usually with the approach of morn- ing, she rolls up the web and some of the adjacent foundation lines, drops the ball to the ground, and moves to her normal daytime resting place against a twig. This intriguing method of capturing prey has no exact counter- part among spiders of any other family, but the device parallels in a general way that of the bolas spider (vide infra). Hackled-Eand Orb Weavers. An outstanding achievement of the Uloboridae has been the invention of an aerial orb web equal in symmetrical beauty and similar in fabrication to that of the ecribellate orb weavers. It is believed by many that this creation is a novelty separately arrived at, not one result of an ancient habit common to allied spiders that later diverged. The germ of an orb web is observable in the great regularity of the webs of hackled- band spinners less versatile than the Uloboridae. Even the irregular mats of Amaurobius and Filistata are based on a framework of dry rays arising from a central retreat. Many species of Dictyna spin aerial sheets of such regularity that in form they approach sectors from the webs of the uloborids. Once a symmetrical design had been realized, lifting the orb 23 C. Ackerman, "On the Spider, Menneus camelus Pocock, Which Con- structs a Moth-catching, Expanding Snare," Annals of the Natal Museum, 1926, p. 418. 150 AMERICAN SPIDERS web from a surface to an aerial station was a relatively simple step. The orb web of Uloborus lies most often horizontal, or slightly inclined, and is only rarely the vertical structure of the typical orb weavers. The horizontal position is a less favorable one, dependent for success on insects that fly upward against it or drop down upon it; whereas the vertical web can intercept much larger flying fauna. Featherfoot Spiders. As has been noted, the curious spiders of the genus Uloborus are most numerous in the tropics of the world, relatively few varieties occurring in the north. Several distinct species are found in the southern United States, but only one, Uloborus americamis, appropriately named the "featherfoot spider" (see Plate XXIV), is common all over the United States and in south- ern Canada. This uloborid has a carapace longer than broad, which is provided in front with eight eyes, in two rows, whose small size confirms the slight reliance this aerial creature places on eyesight. Its chelicerae are moderately robust, but no venom glands are asso- ciated with them— a condition almost unknown in other spiders, and suggesting that the sicky spiral of the orb web and the jaws of the spider are adequate to quiet its prey. The long front legs are often curved, in many species being provided with the tufts of feathery hairs that are the source of their common name. The abdo- men is often surmounted with humps and bedecked with pencils of hairs. A pronounced variation in coloration is characteristic, and pale white, speckled, lined, or all black specimens are often found in the same species. The relatively small orb webs of the featherfoot spider, four or five inches in diameter, are usually placed close to the ground in moist, shaded situations— on low bushes and underbrush, on dead sticks, in hollow stumps, or among rocks. The invariably horizontal web is composed of the same elements as that of the typical orb weavers: foundation lines, radii, dry spiral scaffolding, and a con- centric series of sticky spirals. Before laying down these latter, Uloborus spins a typical preliminary spiral scaffold of dry silk that is used as a bridge to the next radius. The spirals are a composite oi viscid and dry threads, as in the Argiopidae, but here the sticky material is carded from the cribellum with the aid of the very regu- lar row of calamistral hairs on the fourth metatarsus. The spider has never depended on artistic perfection for its capturing snare, but rather on the sticky lines, and close examination shows that the web is imperfect in many respects. Quite often it is most unsym- PLATE XIX a. Male and female, with eggs, in tangled web George Elwood Jenks b. Female holding mass of recently hatched young LONG-LEGGED CELLAR SPIDERS, Pholcus phalangioides George Elwood Jenks PLATE XX Richard L. Cassell a. The spider approaches as the cricket touches the capture threads Richard L. Casstll b. Nooses of swathing film are combed over the leg A COMB-FOOTED SPIDER, THE BLACK WIDOW, Latrodectus mactans, CAPTURES A JERUSALEM CRICKET THE CRIBELLATE SPIDERS 151 metrical— especially during the cocooning season, when the feeding instinct is replaced by a maternal one— and irregular in its details, but it remains quite as pleasing to the eye as the snares of the typical orb weavers. Special features of the web are the hub, which is closely and beautifully meshed, and the ribboned decorations or stabilmenta that ornament the orb and possibly add to its strength. The most frequent form of the stabilmentum is a scalloped band that crosses the central portion of the orb; it is scarcely visible at the delicate hub. Other variations are numerous, a common one being a ribbon coming from a nearby sector to form a V-shaped figure; or four ribbons forming a cross; or broken or completed circles around the hub. In position, the featherfoot spider lies stretched out beneath the hub of her web, her legs directed forward and backward to form a bridge between the stabilmenta and make a complete band across the snare. As she hangs there, swaying with the breeze, she often resembles a bit of leaf or stick. When her eggs are laid, she places the several elongate sacs in a row across the web, and then aligns her long body so that she becomes almost indistinguishable from them— one in a line of bits of debris. Among the tropical uloborids, it is interesting to note, are many social spiders that spin immense webs, where large numbers of males and females live amicably together. These colonial webs are fea- tured by a large central retreat suspended from many long silk lines running in all directions and forming a loose maze. Most of the males, as well as many females and spiderlings, live in the inner part of the web; but from time to time— and this is a particularly fas- cinating part of their activity— individuals detach themselves and move to open spaces on the periphery, there to spin their own characteristic round webs. The outer part of the communal web provides snare-space for all the spiders, and part of the time they live singly in their tiny orbs. Mating takes place in the central re- treat, and the egg laying occurs there as well. Several social species are found in the United States, but their communal webs are rarely notable for size. Uloborus arizonicus, occurring in the Santa Rita and other mountain ranges of south- ern Arizona, will completely invest a low shrub with its web, and several dozen individuals will live there together. Except for size, the colonies of this species closely resemble those of the tropical uloborids. i52 AMERICAN SPIDERS Triangle Spiders. Almost anywhere in the United States may be found the peculiar triangular snare of the spider Hyptiotes, which, because of its small size and retiring habits, is far less familiar than the web it spins. A small creature, rarely more than one sixth of an inch in length, the triangle spider hangs back-downward from a dried twig in its favorite trapping site, and is no more noticeable than a bit of dead wood, a bud, or a piece of bark. The carapace is broad and low; it supports a thick, oval abdomen on which are usually visible slight humps set with a few stiff hairs. Drably clothed in grays and brown, Hyptiotes harmonizes rather well with the dry branches of its home, and affords a striking illustration of close resemblance to environment. All eight eyes are present in this species, but one minute pair is so well hidden in the hair covering that the spider was once thought to have only six eyes. The male ordinarily becomes adult in the early fall, and at that season may sometimes be found near the web of the female, which sex he re- sembles closely except for smaller size. Two well-marked species of Hyptiotes occur in the United States and Canada. The common species in our eastern states is Hyptiotes cavatus, Hentz's triangle spider; while the boreal tri- angle spider, H. gertschi, is abundant in the western part of the country, and largely replaces the other species in eastern Canada, where it occurs as far south as Maine and New York. The web of the triangle spider (Text Fig. 4, B) is best under- stood by comparing it, as did Professor Bert G. Wilder in his early studies of cavatus, to an ordinary pie. The orb of Uloborus is an entire pie; that of Zilla, one of our typical orb weavers, is a pie with a piece cut out of it; and that of Hyptiotes is the missing piece. This triangular web consists of a fifty-to-sixty-degree sector with radii twelve to twenty inches long. It invariably consists of four rays of dry silk, across which are laid down ten or more viscid lines of hackled band that correspond to sections from the spiral line of an orb web. The four rays are attached to an arc line tied to twigs, and converge near a point on a single bridge line fastened to some nearby object. The spinning of the web, often accomplished during the early hours of the evening, is a most interesting process; and the details corroborate the belief that its structure is derived from the uloborid orb web. The first line is a bridge from the resting site of the spider to an adjacent dried twig. It is customary for Hyptiotes to place the bridge line by hand, moving around the periphery of her hunt- THE CRIBELLATE SPIDERS 153 ing grounds to the point of attachment and then pulling the line tight; but in many instances air currents are called upon to balloon the line to a mooring point, as is the practice of the typical orb weavers. A vertical thread from one end of the bridge line is tied to a twig, and forms the arc of the sector. The third principal line returns to near the other end of the bridge line and completes the triangle. Then, between the radial lines, Hyptiotes places two more rays. At this stage the spider has constructed a sector of four rays attached at one end to a single line and at the other to an arc thread. Upon this must now be placed the viscid threads that will make the web a trap. But before the sticky lines are added, Hyptiotes spins a row of three or four dry scaffolding threads, extending from the apex toward the middle of the triangle, that serve to steady the web by holding the radii in place, and that will simplify the laying down of the cribellar silk by providing a bridge from ray to ray. These scaffolding threads are analogous to the dry spirals or spiral bridge of the ecribellate orb weavers, are put down in the same sequence from the apex of the triangle (or hub) outward, and are eliminated in much the same way— bitten out when the web is fin- ished. To lay down the viscid sections, Hyptiotes crawls along the up- permost ray nearly to the point at which it joins the arc line, spins and attaches a band of sticky silk, then crawls back toward the middle of the triangle, spinning as she goes and holding the thread free of the ray. When she reaches the outermost scaffolding thread, she descends upon it to the ray immediately below, and upon this returns, reeling the sticky line back in, until she is immediately be- low the first point of attachment. Here she fixes her line. In order to put down this first vertical cribellar thread, which extends only three inches or so between the two upper rays, Hyptiotes must often crawl forward and backward a dozen inches. One might ask at this point why Hyptiotes does not drop down directly to the ray below? The triangle spider knows her web only by the touch of the silk, cannot see or know the position of the other rays, and is dominated by instinctive actions that keep her pursuing a pre- scribed course. The spider continues this roundabout process until the four rays are bound together by a slightly zigzag vertical line of three sec- tions. She then crawls around the triangle to the top ray once more, and starts on a second line, using her legs to measure its dis- i54 AMERICAN SPIDERS tance from the first. The final number of these partial spirals varies from ten to more than twenty. As the series meets the scaffolding, these latter threads are cut out of the web. The finished web is an extraordinary structure, and is employed in an extraordinary way to provide the spider's food. Hyptiotes takes up a position at the end of the bridge line, near the apex of the triangle, her hind legs touching the silken anchor almost in contact with the twig. With her front legs she pulls the line until the whole web becomes taut; then, holding the slack thus gained over her body, she settles down to wait for her prey. A small moth or other flying insect strikes the web and adheres, struggling violently in the viscid coils. Immediately Hyptiotes lets go of the slack. The web snaps forward, carrying the spider out a short distance with it, and the resultant vibration of the swaying, sticky line causes the victim to become more firmly enmeshed. Hyptiotes seems able to estimate the character of the insect from the nature of its frantic struggles and acts accordingly. The snare may be drawn tight once more and snapped, and this action will be repeated again and again until the spider is ready to crawl over her lines to the victim. Hyptiotes never bites her prey as do many other web spiders, a fact undoubtedly related to the absence of poison glands in this family. Instead, she comes up to the insect, turns her back to it, and, rolling it over and over with her legs, covers it with a thick bluish web. Completely helpless, the victim is carried back to the resting site and sucked dry in the leisurely manner characteristic of the triangle spider. This method of overpowering prey by means of thick bands of silk is analogous to the habits of the comb-footed spiders and the typical orb weavers. Not infrequently, when the trap has been sprung for the first time, Hyptiotes will move forward and, grasping the radii in her front legs and cutting some of the lines, will gradually bundle up the web and hurl sections of it over the victim. By so doing, she destroys the web almost completely, and must spin a new one for her next period of trapping. But inasmuch as one victim provides this small spider with sufficient supply of food for a day or more, the loss of the snare does not materially handicap her. One wonders whether Hyptiotes has not gone to more trouble than the web is worth in producing her triangle trap. Although it will probably catch more insects because of its vertical position and greater size, almost as much spinning and silk goes into its fabri- cation as is expended in the horizontal web of Uloborus. Further- PLATE 2 I W. A. Pluemer Orange Argiope, Argiope aurantia, in web, side view PLATE 22 Edward A. Hill Spiny-bodied spider, Micrathena gracilis, spinning THE CRIBELLATE SPIDERS 155 more, the trick of snapping the trap in order to further enmesh the prey may well be an unnecessary precaution; and the careful enshrouding of the bound victim is likewise an act of doubtful necessity. It is true, however, that the same kind of objection to needless efficiency can be leveled against the snare of the typical orb weavers. The Single-Line Snare. The stick spiders of the genus Miagram- mopes are creatures of the tropics, and although they occur in the West Indies and in Mexico, do not quite reach the subtropical zones in the United States. They must be mentioned here because of the marvelous trapping device they have developed— a device that represents an even greater simplification of the orb web than does the snare of Hyptiotes. The four-rayed triangle is reduced to a single line. The stick spiders resemble Hyptiotes in general structure, but they are more elongate, and are thinly covered with dull grayish hairs over a dusky brown body, so that they almost perfectly re- semble small, thin sticks. On the carapace are four pairs of eyes, the two front pairs being so small and so well hidden that only the hind ones are easy to discern. The front legs are long and thick, stretched forward in close contact with each other; against them presses the short second pair; while the hind legs extend backward along the sides of the abdomen, and fit closely against the body to enhance the remarkable sticklike appearance. The snare of Miagrammopes (Text Fig. 4, A) is a single horizon- tal line, attached at both ends to branches, that stretches about four feet across open spaces in the forest. Conrad Ackerman has de- scribed the activity of spiders on the Natal coast, which, after laying this basic line, then card out a heavy band of viscid silk across its center for a distance of approximately eighteen inches. The next step resembles the triangle spider's method. Miagrammopes moves to the end of her foundation line, and, assuming a position in which she almost touches the mooring twig with her hind legs, appears to be a continuation of it. She draws the line very taut, until she has a loop of slack to hold over her body. The thick center of the snare offers an attractive and familiar-looking resting place for gnats, flies, and a whole host of flying insects. Whenever one alights, the stick spider lets go the loose thread and shoots forward with the elastic line for about half an inch. The release of tension jerks and sways the thread, causing the victim to become more completely 156 AMERICAN SPIDERS entangled. Miagrammopes then rushes to the site of the capture and, again like Hyptiotes, further enswathes the unlucky insect in bluish silk, which she reels and combs from her spinning organs. When her victim is completely helpless, she cuts it loose and holds it in her jaws and palpi. She then closes the rent in the trap, and crawls back to her retreat, where she adjusts the line for the next capture. CHAPTER IX The Aerial Web Spinners IT J.HE VAST ASSEMBLAGE OF SPIDERS treated in this chapter are those that spin silk from their bodies and produce many types of aerial webs. Whereas their relatives developed alertness, speed, brute strength, and a minimum use of silk, to become hunters, the sedentary types on their gossamer lines swung far aside from that line of ecribellate spider evolution. Theirs is a story of silk; on tiny claws that have become increasingly effec- tive as hooks, they hang upside down from the threads of a circum- scribed web, rarely leaving its confines voluntarily. Their sense of sight is rather poor; for this deficiency they have compensated by spinning expansive tangles, sheets, and formal web designs to enlarge their area of action, the struggles of an insect in the farthest recesses of the snare are communicated to them. Within the confines of the web the sedentary spiders have become supreme autocrats. They are a motley crew running to all sizes and shapes. Many are shy, lie immobile in the web, and when disturbed drop on drag- line threads to the security of deep underbrush. Others stay hidden away in retreats or under objects until their traps are touched by small animals. Some are quite agile and run nearly as well as hunt- ing spiders; others, in appearance well proportioned for running, have legs too long and thin. A few specialize in inaction; they hang like inanimate slivers or clods in their webs, to all intents part of the debris that adheres to the lines. Others, possessing greatly elongated abdomens that they wave gently back and forth, resemble in form and action common caterpillars. Most are fat creatures with short legs that seem molded for an acrobatic life. The great majority are tiny and inoffensive; therefore they rarely come to our notice; but some of the orb weavers are giants, in bulk exceeding some large hunters. The webs of the sedentary spiders, displayed on every side in a myriad of sizes and designs, vary from crude artistry to extraordi- '57 158 AMERICAN SPIDERS nary workmanship. Such diverse structures did not come into being at a single stroke; they are the results of long, random experimenta- tion, during which only those suited to the minimum needs of the moment had survival value. From the first wild dragline threads laid down in haphazard fashion on and around the egg sac have evolved by progressive steps the many remarkable snares that today meet the eye. At first there were mere tangles of lines stretched without particular design, roughly filling an allotment of space between suitable supports. Probably altogether composed of dry silk, these mazes were suitable for stopping jumping or flying insects, and retarding their movements through the entangling toils. The addi- tion of viscous drops to the lines was a later development, which transformed the stopping web into an adhesive trap. Among the lines was stationed the egg sac— the central theme, and the theoreti- cal point from which all space webs take their origin. The first spiders that climbed into shrubs were daring adven- turers leaving behind the soil domain so long cherished by their forebears. They could become full-fledged aerial types only after the web novelty had proved its worth as a means of providing food. But once the space web was in reality successful, the incessantly spinning spiders began to explore its possibilities in all directions. Some suspended a horizontal platform of rather loosely woven silk through the middle of the maze and maintained the egg sac at its core. Clinging to the underside of this, they learned to seize insects that, arrested in flight by the maze of threads, would drop to the upper surface of the sheet. The orb web would seem to stand alone as a glorious creation, an incredible novelty designed by superior artisans. That it is only an advanced stage arrived at by the same slow steps that realized the dragline, the stopping maze, and the horizontal platforms is shown in the numerous intermediate examples. The orb web is merely a formal expression of the horizontal platform. Probably at first composed wholly of dry silk, it is now provided with a large area of sticky spirals, and has been swung to a near-vertical position to make it a more effective snare. Almost invariably associated with it are some of the lines that were once the stopping maze. The space webs exhibit a most interesting evolutionary series. Each major web type has been sponsored by different groups of aerial spiders. The primitive line weavers still rely largely on the tangle of threads for protection and as a means of stopping their prey. The comb-footed spiders spin a maze, sometimes a sheet, and THE AERIAL WEB SPINNERS 159 almost invariably fix guy lines with sticky globules to hold their vic- tims. The sheet web weavers use both maze and sheets of various forms. The sticky spirals of the orb weavers hold fast an array of jumping and flying insects. Along with the webs, there have de- veloped most interesting techniques for overpowering and enmesh- ing the victims; and the basic factor upon which these techniques depend is the spider's ability to move upon the web. The successful venture in silken lines is made possible by the unpaired median claws, which lie between the much longer outer pair and near their base. The median claw in aerial spiders is shaped like a hook, and is provided with a few small teeth. Associ- ated with it are various modified hairs, which, often curved and toothed, are called "spurious" or "accessory" claws. The median claws are used almost exclusively for clinging to the lines of dry silk. They are displaced slightly to the sides, those of the first and second tarsi toward the anterior and those of the third and fourth tarsi toward the posterior outer claw. This facilitates grasping of the threads, which fit into the hook of the claw without requiring a turning of the tarsus. When walking in the web, the spider draws the tarsi across the threads to catch the median claw, which grasps the line at an acute angle and twists it to make the grip firmer. The spurious claws orient the threads so that they can be hooked by the median claw, then act to clear the thread from the notch by uncoupling and hurling it out. With this effective device, the aerial spider moves through deep mazes or across vertical meshes with ease and precision. Before passing on to brief sketches of the major groups of aerial spiders, some generalizations can be made that indicate the pro- found differences between them and the hunting spiders (discussed in the following chapter). Both are derived from the same prim- itive stocks, and on their separate roads both have become amaz- ingly specialized. The success of each line is attested by the vast number of species found living side by side, and by the develop- ment in each series of a wide and amazing variety of types. In terms of degree of change from prototypes, the sedentary spiders have outdistanced the hunters. The spinners live in holes under the ground, they live near or on the surface, they live in surface vegetation, shrubs, and high up in trees. They have invaded aerial space with their threads, and claim it as their own by mere place- ment of their three-dimensional webs. As for the vagrants, they are dominant on the soil and in the various strata of plants. They 160 AMERICAN SPIDERS often claim space beneath the soil by digging a tunnel; and the water spider has invaded the fresh water with great success. Above the ground, the vagrants move on all types of surfaces and climb into shrubbery with great agility. It may be seen, therefore, that they live side by side with the aerial spiders, but both are neverthe- less to a large extent insulated from each other, almost as if they were in two worlds. King in its own domain, the hunter is usually a weakling in the clutching web of the sedentary spider. Outside its web, the sedentary spider is no fair match for the average hun- ter. The superiority of either line can never be tested except in terms of which one shall give rise to the dominant spider of the future. Both have accomplished great things, and stand as equals that have reached their goals by different roads. THE PRIMITIVE LINE AND SHEET WEAVERS The members of this group can be regarded as a major segment of that series that took to an aerial life. They resemble in general features and equal in developmental rank the primitive hunters and weavers. For the most part, they are pale spiders that live in dark places, there laying down a relatively simple web of dry lines or sheets and relying on this to secure their livelihood. Most are little changed from the presumed ancestral types. The palpi and epigyna are quite simple, though in one family, the Pholcidae, they appear to be specialized by numerous processes that largely mask the other- wise generalized nature of the organs. The posterior respiratory organs are tracheae. In the Telemidae are two openings to the tracheal tubes, but in the other families only a single one is present, the usual position being well in front of the spinnerets. The chelic- erae of the Pholcidae are soldered together along the midline as in the Scytodidae and related families, but in the other members of the series they are free. Males and females are quite similar in size and appearance; often they are found living amicably together in the webs. During the mating, the pholcids insert both of the palpi simultaneously, as do most of the primitive hunters, and the stance is the generalized type of that group. Little is known about the mating habits of other members of the series. The line weavers of the family Pholcidae (Plates VII and XIX) have small globose or elongate bodies suported on exceedingly long and thin legs, a physical feature that causes them to be mistaken for THE AERIAL WEB SPINNERS 161 the daddy longlegs. The leg tarsi are often made flexible by the presence of numerous transverse creases or sutures in the integu- ment. Eight eyes, set close together on an elevated tubercle, are usually present, but the anterior median pair may be lost, and in some cave species all eyes may be reduced in size or completely missing. The long-legged pholcids occupy a position between the higher sedentary types and the spiders of very primitive level. Their derivation from prototypes similar to the present cribellate Filistatidae through loss of the cribellum and modification of a few other features is quite plausible. Although commonest in warmer regions, they are quite numerous in temperate areas, as is well shown by the presence of eight genera and about forty species in the North American fauna. The pholcids spin, in dark places, loose, irregular webs, some- times with a distinct closely woven sheet. Males live in the same webs as the females and resemble them closely, but may be recog- nized by the great size of the palpi, which are enlarged to form thickened appendages. The females carry the eggs in their chelic- erae, glued together into a spherical ball and tied lightly with a few silken lines; later they may be found holding the mass of recently hatched young. Most pholcids are pale white or yellow, but some are more gaily colored in pastel greens and blues. Many become domestic, especially in our southwestern states, where species of several genera find conditions in houses and buildings quite as suit- able as in the open. These long-legged line weavers are like some of the orb weavers in having a most interesting habit that becomes operative when prospective insect prey is caught in the net. They shake the web violently to hasten thorough entanglement, then, when the capture is being made, twist the victim around and swathe it with silk. This aggressive action turns into a defensive gesture when the spider is disturbed, and it pumps up and down on its long legs so violently that it becomes a mere blur. This whirling or shuttling, which be- comes increasingly violent when the stimulus is repeated three or four times, usually takes place when the web or the body of the spider is touched, but on occasion other stimuli provoke the re- sponse. When thoroughly aroused, the spider retreats to dark re- cesses within the web, or drops down from it to run rapidly and hide away in some dark corner. The best-known member of the family is the long-legged cellar spider, Pholcus phalangioides (Plate XIX), which occurs in houses i6i AMERICAN SPIDERS almost everywhere in the world. A relatively large creature with a pale white, elongate body a quarter of an inch long and legs two inches long, it covers the ceilings and walls of our cellars and neglected rooms with its maze of cobweb. In this section mention may be made of three families of prim- itive line weavers that differ from the pholcids in having six as the normal number of eyes. All are tiny creatures, rarely more than an eighth of an inch in length, which live retiring lives in dark places under stones and debris on the ground or in caves. The relatively small number of species known from North America reflects a failure to explore our caves adequately, rather than any true sparsity of these minute animals. The sheet weavers of the family Leptonetidae have relatively slender bodies and fine, long legs. The eyes form a V-shaped figure, four close together in front, the posterior median pair set quite far back. The web is an extensive sheet of finely spun tissue placed in fissures on cave walls, and has no definite maze of lines associated with it. Tiny white egg sacs containing few eggs are placed on the walls near the web or hung from the web itself by a thread. Several species are known from the southern portions of the United States. Closely allied to the leptonetids are some tiny cave spiders of Europe and Africa, the Telemidae, which lack book lungs and have four orifices leading to tracheal spiracles. Telema tenella of the eastern Pyrenees is eyeless. In the western United States occur representatives of another fam- ily of these primitive spiders, the Ochyroceratidae, which have tiny globose abdomens much like those of the pholcids. The six eyes are placed in a transverse row across the front of the head. A typical species is Usofila gracilis, which, only one twenty-fifth of an inch in length, occurs in Alabaster Cave, California; other species live under debris on the soil outside caves. THE COMB-FOOTED SPIDERS The comb-footed spiders of the family Theridiidae are for the most part thickset sedentary types that hang upside down from the dry threads of irregular maze webs. Most are small spiders, suspending their snares on plants with lines so fine that they are often unnoticed, or hiding them in burrows or fissures in the soil and under debris. Less well hidden are the webs of drab, house- loving Theridion tepidariorum, which, soon covered with dust and THE AERIAL WEB SPINNERS 163 debris, form the cobweb anathema of the neat housewife. One of the most handsome and colorful members of the family is the black widow, whose beauty, however, is marred by its unsavory reputa- tion. A few theridiids have hard bodies ornamented with curious spines; in others the abdomen is drawn out to amazing lengths. Most are inveterate spinners, but a few curious types (Conopistha) live in the webs of other spiders as commensals, and another group (notably Euryopis) has forsaken a formal web for an errant life. Most of the theridiids have rather soft, light-colored abdomens, oval or globose in form, and long, slender legs that lack spines. One of their special features is the presence, on the tarsi of the fourth pair of legs, of a line of enlarged, curved, and toothed setae that form a distinct comb used to fling silk over the prey. In most of the comb-footed spiders, the comb is strong and distinct, but in the smallest ones it may be difficult to see, and in some others it has become reduced to a few modified setae. Their relatively small eyes are set close together in a group near the front of the head. Sight enters their lives only to a limited degree, since they live in dark places and become active chiefly at night. Some males are mere pygmies beside their bulky mates, and there is often a marked sexual dimorphism. The theridiids occur in great numbers in the temperate and tropical zones; within the United States and Canada several hundred different species representing about twenty-five genera are found. Thus mention can be made of only a few that typify the group, or are outstanding for peculiarities of habit. The snare of the comb-footed spiders (Text Fig. 5, B) is not the simple mass of irregular lines that casual study would seem to indi- cate. It has incorporated into its limits some interesting innovations. A densely woven sheet of silk is often a feature, serving as a shelter under which the spider retreats. Leaves and debris, or grains of sand, may be used as building materials. One of the most interesting homes is the bowl of the boreal Theridion zelotypum. Composed of dried spruce needles or other plant parts sewed together with silk, it provides a strong, waterproof tent beneath which the spider can hide with its eggs and young. In some instances the theridiids leave their spherical egg sacs suspended in the scaffolding of lines in plain sight. On the outskirts of the web at the proper season may be seen the mature males, which are received for the most part with kind- ness during courtship and following mating. Males are killed oc- casionally, but not with the regularity ascribed by popular belief. 164 AMERICAN SPIDERS The recently hatched young remain with the mother for some time, and receive consideration far beyond what one might expect from simply organized creatures. The common Theridion notatum of Europe and no doubt similar spiders from many other parts of the world feed their young for several days by regurgitating fluid upon which the babies make their first meals. Thereafter for several weeks, the mother and babies feed together upon insects caught and dragged to the retreat. A typical theridiid web has, in addititon to a central maze with or without the retreat, a series of longer guy lines that anchor the whole against supporting surfaces. These guy lines are held taut near their base by inconspicuous studs of viscous silk. Small insects walking against the lines are held by the glue; when their struggling breaks a line, they are lifted bodily by its contraction. The dis- turbance quickly brings the spider to the spot, and the size of the intruder determines to some extent the reaction. A pictorial story of the technique used by a black widow spider to subdue a large wingless Jerusalem cricket (Stenopelmatus) is shown in Plates XX and XXI and Plate 6. The spider approaches cautiously, no doubt forewarned of the size of the prey by the strength of the pulls on the lines, then turns completely around to present its long hind legs to the victim. With the aid of the comb on its flailing hind legs, it draws out heavy lines of sticky silk and ties them to the leg of the insect, until a strong band is formed. The spider next turns and injects its venom by piercing the leg with its tiny, sharp chelicerae. (Ordinarily the victim is not closely approached until completely fettered.) Then begins the task of lifting the still struggling insect off the snare floor and moving it to a suitable point in the maze. By numerous small steps, during which various threads are tightened and others put down, the bulky vic- tim is hoisted gradually into the air until it is about three inches from the floor. Now follows the banquet. The spider feasts lei- surely for three or four days upon the body of the tightly bound prey; then the much shrunken remains, sucked dry, are gradually lowered beyond the inner maze and dropped to the ground. The theridiids have long been noted for their engineering skill in lifting objects of great size. Common domestic Theridion tepi- dariorum is credited with having overcome and lifted small snakes, mice, and other animals. After presenting the details of captures of various small snakes by this spider, McCook has the following to say: THE AERIAL WEB SPINNERS 165 It is worthy of mention, in connection with these incidents, that the belief that a special enmity exists between spiders and serpents is very ancient. Pliny says that the spider, poised in its web, will throw itself upon the head of a serpent as it is stretched beneath the shade of a tree, and with its bite will pierce its brain. Such is the shock that the creature will hiss from time to time and then, seized with vertigo, will coil round and round, but finds itself unable to take flight or even to break the web in which it is entangled. This scene, concludes the author, only ends with the serpent's death.24 One of the more spectacular feats of Theridion reported by McCook was the subduing of a small mouse: A very curious and interesting spectacle was to be seen Mon- day afternoon in the office of Mr. P. C. Cleaver's livery stable in this city. Against the wall of the room stands a tolerably tall desk, and under this was a small spider, not larger than a com- mon pea, who had constructed an extensive web reaching to the floor. About half-past eleven o'clock, Monday forenoon, it was observed that the spider had ensnared a young mouse by passing filaments of her web around its tail. When first seen the mouse had its fore feet on the floor and could barely touch the floor with its hind feet. The spider was full of business, running up and down the line and occasionally biting the mouse's tail, mak- ing it struggle desperately. Its efforts to escape were all unavailing, as the slender fila- ments about its tail were too strong for it to break. In a short time it was seen that the spider was slowly hoisting its victim into the air. By two o'clock in the afternoon the mouse could barely touch the floor with its fore feet; by dark the point of its nose was an inch above the floor. At nine o'clock at night the mouse was still alive, but made no 'sign except when the spider descended and bit its tail. At this time it was an inch and a half from the floor. Yesterday morning the mouse was dead, and hung three inches from the floor. The news of the novel sight soon became circulated, and hundreds of people visited the stable to witness 34 Pliny, Natural History, Chap. X, p. 95, quoted in H. C. McCook, Ameri- can Spiders and Their Spinningivork, Vol. I (1889), pp. 241-2. 166 AMERICAN SPIDERS it. The mouse was a small one, measuring about one and a half inches from the point of its nose to the root of the tail.25 This spectacle, watched with amazement by many people and interrupted by the clumsiness of a "meddlesome boy" who acci- dently broke the web (instead of by the intervention of the S.P.C.A., as, is usually the case), is a compliment to the strength and elasticity of the multiple threads of the line weavers, and to their engineering prowess in elevating tremendous loads by block-and- tackle methods. A high percentage of our comb-footed spiders belongs in the genus Theridion, perhaps the largest of all spider genera and typ- ified by Theridion tepidariorum. Most other species are smaller and more brightly colored. The globose female of Theridion differ- ens, one-eighth inch long with a reddish brown abdomen marked above by a red, yellow-edged stripe, places her large white egg sac in the nest. Her web is found on low plants of all kinds, and con- sists of a small tent, barely covering the spider, from which an ir- regular network of lines spreads out across the limits of the plant. Representative of another species group is Theridion frondeum. This spider has a pale white or yellow body boldly marked with black, but extremely variable in color and pattern. Some examples are almost entirely white, unmarked, whereas others have narrow dark lines or bands on the cephalothorax, and small black spots, dusky bands, or dark stripes and patches on the abdomen. These handsome theridiids, represented by one or more species almost everywhere in the United States, live on low plants and prefer moist, lightly shaded areas in woods or along streams. Closely allied to the theridiids are the species of Tidarren, the best known of which, fordum, resembles tepidariorum in size and coloration and lives in similar situations. Whereas the males of Theridion are inferior to the females in size— a disparity reflected chiefly in the lesser bulk of the abdomen, the males of Tidarren are babies by comparison. The female Tidarren fordum is often nearly one third of an inch in body length, whereas the males are rarely larger than one eighteenth of an inch. At the proper season these pygmies often cluster in the webs of the females, usually a dozen or more to a web and rarely fewer than two or three, and seem to be tolerated. The males of the known species of this genus carry ^McCook, ibid. THE AERIAL WEB SPINNERS 167 only a single palpus, a large bulbous affair, the mate of which is extirpated just before full maturity. Remarkable for their social habits are the species of Anelosimus, close relatives of the theridions but having more elongate bodies. Our single well known species, studiosus, is a light brown spider one sixth of an inch long with dark upper and lower stripes on the abdomen. It is abundant in the South, and occurs as far north as New Jersey. Its communal web is placed on shrugs and trees, and ordinarily comprises an unsightly mass of dead leaves tied together with silk and serving as a retreat, around which extends a sheet of silk attached to twigs. Several individuals live together in the nests, which, except for size, are like those of other gregarious species. A very similar species abundant in Brazil, Venezuela, and Panama is Anelosimus eximius, once dubbed socialis and well known for its social habits. Colonies of hundreds or thousands of individuals, males and females and immature stages, spin a light, transparent web, similar in texture to the sheets of the grass spiders, which has little definite form and may completely invest sizable shrubs and even trees. Some are a yard across, and are spun fourteen or fifteen feet up into the foliage of trees. The spiders wander about freely within these confines, and feed communally on insects that are captured at the periphery of the web and carried into the interior. Sometimes found in the webs are such vagrant spiders as Sergiolus and the two-eyed Nops; they may be predators or perhaps sym- bionts, but their exact relationship to the aggregation is not known. One group of theridiids deserves mention both for curious body forms and for commensal habits. These spiders are mostly small, and, except for the vermiform types, rarely exceed a third of an inch in length. Their legs are long and very unequal, and the tarsal claws are remarkable in that the unpaired one is long, only slightly curved, and may actually exceed the paired claws in length. The tarsal comb is reduced to three or four modified setae. Both the cephalothorax and abdomen are subject to curious variations within the three known genera. In Ariamnes, the abdomen is drawn out into a long and slender cylinder that ends in a point; in Rhomphaea, it is usually triangular in shape, sometimes extremely high, and oc- casionally vermiform as in the preceding genus; in Conopistha, the abdomen takes many forms, being spherical, triangular, or cylin- drical, and embellished with lumpy or pointed projections. In both Rhomphaea and Conopistha the heads of the males are ornamented 168 AMERICAN SPIDERS with rounded lobes, protruding trunks, elevated spines, or other curious processes, some of which may bear the eyes. Our single species of Rhomphaea is fictilia, a silvery spider with darker bands on the cephalothorax and a single band down the middle of the abdomen. It occurs all over the United States. The body, variable in length and about a quarter-inch long, is somewhat triangular, but in some examples may be drawn out into a vermi- form appendage half an inch long. This long-legged spider spins a tiny web between leaves or blades of grass, where it hangs like a straw. Its egg sac is a yellowish object shaped like a slender vase and about the same size as the spider. Fictilia closely resembles some of the more typical wormlike spiders of the tropical genus Ariamnes, and like them is able to bend its elongated abdomen back and forth. Regarding this appendage, F. O. P. Cambridge has the following to say: This, as I have myself observed in Brazil, is wriggled to and fro, looking like a small caterpillar. But of what service to the spider this accomplishment may be is not easy to guess; for on the one hand it seems likely to attract the attention of grub- eating wasps and ants, though on the other it may attract, within striking distance, gnats and small flies who become curious to ascertain what the wriggling phenomenon may portend.26 The best-known genus is Conopistha, which comprises the multi- tude of commensal types heretofore known by the generic name Argyrodes, of which quite a number of species occur in the United States. All are known to spin tiny webs of their own, but they are more frequently found hanging in the webs of orb weavers, line weavers, sheet weavers, and not uncommonly in the snares of grass spiders. While hanging in these webs, legs closely drawn together against their bodies, they present an amazing resemblance to straws, twigs, scales, bits of leaves, and debris, so camouflaged that they are completely lost except to the most practiced observer. Largely immune to attack from their hosts because of small size, and per- haps also because of their cautious movement within the lines (in limited sectors of which they lay down threads of their own), they feed upon the tiny insects disregarded by the host. One of our commonest species is Conopistha trigona, a yellow- ish, triangular spider scarcely an eighth of an inch long. The ab- 88 Quoted by J. H. Comstock, in The Spider Book (1940 ed.), p. 352. THE AERIAL WEB SPINNERS 169 domen is high and pointed. The head of the male is ornamented with a rounded horn between the eyes and another, more slender, just in front, but that of the female remains normal. A related species is Conopistha nephilae, a pretty black and silver spider abundant in the South that favors the webs of the larger orb weav- ers, notably those of the silk spider Nephila. The head of the male is produced in two long lobes, of which the upper one bears the four median eyes. A close relative called pluto lives in the webs of the black widow spider in northwestern Mexico. Another common species, Conopistha cancellata, has an elongate, triangular, gray or brownish body marked with a few silver spots and set with paired lobes on the side and at the end of the abdomen. The head of the male is produced in a rounded lobe on the clypeus, above which are two pits. This spider resembles a piece of bark or a dead leaf when lying in the web of an orb weaver or of a grass spider. Typical comb-footed weavers place complete reliance on a maze of dry lines, sticky droplets, and films from the lobed glands to ensnare insects. A few theridiids, on the other hand, have been able to divorce themselves from silk as the only means to capture prey. These spiders are small, comparatively flattened types, with legs of moderate length. They live under stones, in moss and leaf- mold, and move over the soil and vegetation with great speed. Little is known about them, but they seemingly hunt their prey as do the hunting spiders, and spin no formal webs. At least two genera from North American fauna, Stemmops and Euryopis, be- long to this series, but mention will be made only of the latter group, which is widely distributed and represented by numerous species. The species of Euryopis, which resemble in a superficial way some of the crab spiders, have heart-shaped abdomens pointed be- hind and covered with a dorsal shield set with long setae. Our commonest eastern species is Euryopis funebris, a handsome black- ish spider one-eighth inch long, whose abdomen is bordered with a silvery white stripe. Several species of similar pattern occur in the South, and prominently in the western part of our country. An- other series of species, which includes Euryopis argentea, is nearly black and has the abdomen pointed with four to six pairs of small silvery white spots. A very similar species is Euryopis spinigerus, orange or brown with a more distinct dorsal shield and more con- spicuous curved bristles. The spiders of the genus Hadrotarsus, known from the Aus- i;o AMERICAN SPIDERS tralian region and ordinarily placed near the Oonopidae, are sed- entary types similar to Enryopis that have become vagrant secondarily. Although they have lost their unpaired claws, they still retain spurious claws and numerous other features that point to an origin from the comb-footed spiders. THE SHEET WEB WEAVERS The addition of a formal horizontal platform marked a signif- icant departure from the irregularity of the tangled space web. This strengthened zone of thin and loose webbing, with the egg sac at its hub, quickly became the theme of a new type of snare in which the upper and lower mazes and the guy lines now played a subsidiary role. The germ of the platform was present in the webs of some of the comb-footed spiders, but developed no further there; the closely woven sheet of the sheet web weavers and the geometric snare of the orb weavers, however, are its direct results. These latter spiders represent a common stock that, though early branching onto separate roads, has come down to modern times as two closely allied lines. So much in common have these dominant aerial spinners that they were for a long time classified within the limits of a single large family. The comb-footed spiders diverged from the line at nearly the same time, perhaps because of failure to introduce regularity into their web by exploiting the platform, and took a path toward perfection of lobed glands and tarsal combs. The sheet is a yielding table upon which drop flying and jump- ing insects, usually after being halted in midair by a superstructure of crisscrossed lines guyed to adjacent vegetation. The sheet web weaver clings upside down beneath the blanket, runs over the sur- face with rapidity, and pulls its prey through the webbing. The principal sheet acts as an effective screen against enemies from above, as well as a relatively efficient snare. A second sheet is often present beneath the hanging spider, apparently serving as a barrier to attack from below. The sheet webs (Text Fig. 5, A) are used for a long time; when partially destroyed by winds or falling debris, they are replaced after a few hours of spinning. In some instances, the stopping maze above the trap is missing, or is represented only by a few guy lines. Snares placed near the ground are effective in stopping and holding collembolans and small insects of many types. The sheet web weavers of the family Linyphiidae far exceed in PLATE 23 J. M. Hollisler a. Shamrock orb weaver, Aranea trifolium J. M. HaMstcr b. The garden spider, Aranea diadema J. M. Hollisttr c. Orb weaver, Neoscona }. M. Hollister d. Orb weaver, Neoscona, on leaf ORB WEAVERS PLATE 24 Walker Van Riper, Colorado Museum of Natural History Wolf spider, Geolycosa missouriensis, at mouth of burrow TEXT FIG. 5.-WEB TRAPS OF AERIAL SPIDERS (SCHEMATIC) A. Maze and sheet web of Linyphia. B. Maze and capturing lines of Therid- ion. C. Casting line and globule of Mastophoi a. D. Maze and domed orb web of Allepeira. E. Maze and orb web of Metepeira. THE AERIAL WEB SPINNERS 171 numbers of genera and species the total for any comparable group in the temperate zones; they are the dominant aerial types. Most of the species are small, even minute, and they occur in vast, little- noticed numbers under soil debris. As a group they have more elongated bodies than the comb-footed spiders, and their legs are set with long spines. Few of them become the obese lumps so fre- quently found among the orb weavers; many run over the soil with a speed that belies their dependence on a fixed space web. Their chelicerae are large, strong, and well-toothed, the straight maxillae little if at all inclined over the labium. The presence of stridulating organs, most frequently a file on the side of the chelicerae and a scraping spine on the femur of the palpus, further differentiate them from the orb weavers. Sexual dimorphism is not pronounced, except in the species with modified heads, and there is often considerable similarity in size and coloration of the sexes. Male and female live peaceably together in the webs during the summer months. Most linyphiids are rather plainly colored, but there are no- table exceptions; some are carmine red (Ceratinopsis), and many have distinctive dark patterns on light bodies. As is true of most sedentary spiders, especially of those that live in dark situations, the eyes are small, and little used, if at all, for the location and capture of prey. Linyphiids for the most part prefer the shade, conse- quently they live hidden away in dark places under natural debris on the ground, or beneath the leaves of living trees. Many species dwell in caves or animal burrows, and have in certain instances partially or completely lost the eyes. The linyphiids are divided into two principal groups, which, quite distinct in their extremes but completely bridged by inter- mediate forms, are placed by many araneologists in separate fam- ilies. The first of these, the Linyphiinae, includes the largest species, as well as numerous spinners of extraordinarily beautiful webs. As regards physical characteristics, the pedipalp of the female usually retains the tarsal claw, and the palpus of the male lacks tibial apoph- yses. In general the legs are longer and thinner, and are set with more numerous spines than those of the Erigoninae (the other principal group); and the tibiae are almost always furnished with dorsal and lateral spines. Many small species belong in this group, but space allows mention of only a few of the larger representatives. One of our largest and best-known linyphiids is the filmy dome spider, Linyphia marginata, which abounds in temperate North 172 AMERICAN SPIDERS America, and is also common in Europe. An elongate spider meas- uring one sixth of an inch, the adult female has a dusky cephalo- thorax with a paler marginal stripe, and a whitish abdomen heavily marked with dark bands and stripes. Marginata's conspicuous webs are often placed along paths or streams in shady, moist woods. The outstanding delicacy and beauty of the snare are fully revealed when the rays of the sun strike it. There is a maze of threads, ex- tending in all directions and tied to adjacent vegetation, at the center of which is a domelike sheet three to five inches in diameter. The spider hangs below the apex of this dome. Flying insects strike the highest lines of the superstructure, drop among the closer threads, then upon the dome itself. There they are greeted by the spider, which pulls them through the webbing, trussing them up with additional silk lines while making the capture and afterward repairing the rent in the sheet. Sometimes the web is shaken to hasten the dropping of the prey. The lines of the maze and the sheet are slightly viscous, but the drops do not gather into the sticky globules used by the orb weavers and comb-footed spiders. The common bowl and doily spider, Frontinella communis, found almost everywhere in temperate and tropical North Amer- ica, is quite similar in appearance to the filmy dome spider. It spins two separate sheets in its snare, the principal one shaped like a shal- low bowl, under which the spider hangs, and the second one a horizontal sheet placed below the spider. A stopping web, largely filling the bowl and extending above it, is tied to twigs of low bushes. Snares characterized by a secondary sheet are spun by various other members of the group. Many smaller linyphiids tie a flat platform web among low plants and move about over the lower surface, but they drop to the ground and run away when disturbed. A few of the linyphiids spin no web and have become errant types. Drapetisca alteranda is the best-known American type. This spider is commonly observed sitting flat against tree trunks, where it pursues its prey and around which it scurries when menaced. Its mottled gray and white body closely resembles the bark of aspens, birches, and beeches, on all of which the spider may be found; against such a background it is difficult to distinguish. The second principal group of linyphiid spiders, the Erigonmae, or dwarf spiders, consists of small spinners that live obscure lives under debris. The pedipalps of the females usually lack tarsal claws, while those of the males are armed with tibial apophyses. Most are shorter-legged than their relatives, and live closer to the soil, run- PLATE XXI Richard L. Cassell c. Tiny fangs inject the venom d. The bulky insect is lifted above the floor Richard L. Cassell A COMB-FOOTED SPIDER, THE BLACK WIDOW, Latrodectus mactans, CAPTURES A JERUSALEM CRICKET PLATE XXII Lee Passmore A female humped orb weaver, Aranea gemmoides, clinging to a plant Walker Van Riper A female humped orb weaver, Aranea gemmoides, hanging in the hub of her orb web HH Edwin Way Teale A fisher spider, Pisaurina mira, with egg sac THE AERIAL WEB SPINNERS 173 ning over it quite actively when shaken from their tiny webs. They come to light chiefly when leaves, moss and organic debris are sifted over a sheet by the careful collector. The erigonids are well known for their aeronautic habits; in autumn they constitute a large part of the total group of fliers. Since most of them are less than one tenth of an inch long, they can fly in adulthood as well as in the younger stages. The small descriptive information given these spiders reflects an incomplete knowledge of their habits rather than their importance, since they are represented by a large number of genera and species. A high percentage of the spiders of the northern hemisphere, as well as most of the hardy boreal types that penetrate far into the cold north and frequent the tops of our highest mountains, belongs to this group. Their tiny flat webs, fortified with a dense covering of viscid droplets, must reap a tremendous harvest of tiny insects to maintain such a population. Some of the best-known members of this series belong in the genus Erigone, which includes numerous dark brown or black spi- ders with smooth and shining carapaces armed on the sides with heavy teeth. The chelicerae and the pedipalpi are likewise often studded with sharp spines. These erigonids are frequently found along the edges of streams or lakes, where they place two-inch- square webs among the grass roots or suspend them across stems over the water. Many male erigonids have heads pitted and modified into gro- tesque shapes. A slender horn, somewhat thickened at the end and set with rows of stiff hairs, extends forward between the eyes of Cornicularia. In Gnathonargus unicorn a single, long, slender horn projects from the middle of the clypeus. A rounded lobe carries the posterior median eyes of Hypselistes florens, and of many similar species, high above the remaining pairs. One of the most amazing of all erigonids is the European Walckenaera acuminata, whose eyes sit in two groups at the top and middle of a slender tower more than twice the height of the head itself. Often associated with these bizarre modifications are deep, conical pits usually placed just back of the posterior lateral eyes. The use to which such pits are put ap- pears to be known only for the European Hypomma bituberculata. W. S. Bristowe noted that during mating the female seized the male by the head and inserted the claws of her chelicerae into the rounded pits. This observation suggests that many other species with pitted heads may perform in a similar manner, and further, i74 AMERICAN SPIDERS that the head modifications, even though pits are absent, may be associated with interesting copulatory routines. The conclusion that modified heads and pits have arisen quite recently and inde- pendently in various groups of erigonids is supported by the close relationship with species that do not exhibit these secondary sexual characters. Among the most interesting spiders of this subfamily are some that have become typical cave forms. Anthrobia mawmouthia, found in the Mammoth and other caves in Kentucky, has lost all traces of eyes. It lives under stones in the cave, there spinning small, flattened egg sacs that contain a few unusually large eggs. Another tiny species, Phanetta subterranea, is a characteristic feature of cave systems from Indiana and Kentucky to Virginia. All its eyes are usually present, but frequently they are much reduced in size, and occasionally the anterior median pair is missing. The cave spiders of the subfamily Nesticinae resemble the theridiids in appearance, and have a somewhat similar comb of toothed bristles on the hind tarsi, but their mouth parts and genital organs ally them with the series of sheet web spinners. Their webs are loosely meshed sheets and tangles hung on the walls of caves or hidden under stones. The females drag globose egg sacs around with them, attached to the spinnerets in wolf-spider fashion. The nesticids always live in dark situations, evidencing a decided predi- lection for caves, mines, and tunnels. They are pale spiders; their eyes are reduced in size or missing; and their allegiance to cave life is reflected in their loss of pigment. Several different nesticid types occur in the United States. A darkly marked species, Nesticus cellulanus (introduced from Eu- rope), lives in cellars and in dark corners in houses and barns. Our common Nesticus pallidus, a pale yellow spider, one seventh of an inch long, found all over North America, lives under stones or boards on the ground, in burrows and cave entrances, and deep in totally dark caverns. Outdoors specimens all have normal, well- pigmented eyes, whereas some cave dwellers have lost the anterior median pair. The pirate spiders (family Mimetidae) are curious aerial types that creep into the webs of other spiders and kill them. These hand- some cannibals, so far as is known, feed exclusively upon other spiders and never use silk for a snare. Their bodies are delicately marked with dark lines and spots. A principal feature is the pres- ence of a series of very long, regularly spaced spines, with smaller THE AERIAL WEB SPINNERS 175 spines between, forming a rake on the metatarsi and tarsi of the front pairs of legs. Other structural details would seem to ally the pirates either with the sheet-weaving linyphiids or the orb weavers, but their virtual failure to use silk in any way keeps their position obscure. In many respects they resemble the enigmatic Archaeidae from Baltic amber deposits, modern species of which have been dis- covered in South Africa, Australia, New Zealand, and southern South America. About a dozen mimetids occur in the United States. Typical situations are ground debris, vegetation, and, of course, the webs of other spiders. The species of Mimetus are about one fourth of an inch long, and have rounded abdomens with two angled humps above the base. The species of Ero are about half as large, shaped much like Mimetus, and have small humps on the top of the ab- domen, with a covering of stiff brown hairs. The egg sacs of Ero are spherical bags covered with a loose network of brownish silk; this is twisted to form a thread by which the bag is suspended above the ground. All the mimetids are slow-moving, stealthy cannibals that have become experts in their nefarious trade. Mimetus preys on the orb weavers and the comb-footed spiders, and in the South is frequently found in the webs of Tidarren fordum. Ero attacks and subdues its prey with an expertness that belies the animal's seeming innocence. This small pirate will craftily enter the tangled lines of TheruKotfs web, and clear a space of threads without making its presence known to the occupant. When all is prepared, Ero pulls at the lines, then awaits the approach of the aroused spinner, which hur- ries to the spot with customary confidence. At just the right mo- ment, Ero grasps the legs and body of Theridion with its long front legs, and, holding on firmly with the coarse rake of spines, quickly bites the femur of the victim's front leg. A complete collapse of Theridion, the consequence of a remarkably virulent venom, is al- most instantaneous, and the victor immediately begins sucking the body juices from the bulky prey. Only on rare occasions are the tables turned, and the pirate made a victim of its own seduction. THE ORB WEAVERS The two-dimensional snare known as the orb web is a crowning achievement of the aerial spiders. To the layman the web is an i76 AMERICAN SPIDERS engineering triumph, a fixed geometrical object that symbolizes spider and partially allays unreasoning distrust of the creature. To poets of all times, the orb, divorced from the spinner itself, is a celestial creation founded on beauty, its graceful spirals symbolic of the heavens and its mystery, its fragile lines a measure of the evanescence of life. To the evolutionist, it is only the last step of a series that has resulted in a circular design— an inevitable shape; and the spider has no more to do with spinning such a symmetrical web than "a crystal has with being regular." The orb web, among all objects produced by lesser creatures an unrivaled masterpiece, is above everything a superb snare. Contemplating it, one echoes the words of the meditative Fabre: "What refinement of art for a mess of Flies!" The orb web, quickly strung up and as quickly replaced when defective, brings to the trapper an abundance of the choicest flying insects. It exploits a food supply that is active both by day and night, and, in the adult winged state, available only by chance to other aerial spinners. Almost invisible in ordinary light, the lines stretch across space as a tough but yielding net, into which fliers blunder, to be held by sticky, elastic threads that make the most powerful wings ineffective. (That a similar trap, produced by a like series of instinctive actions, should have evolved among a sep- arate line of spiders might well seem an impossibility. Nevertheless, the cribellate uloborids have fashioned a web that, except for sub- stitution of the hackled band for the beaded spiral lines, is a faithful reproduction of the snare of the orb weavers.) This most highly evolved of all aerial webs is the result of the random activities of aerial prototypes, which finally established order among the irregular lines in the horizontal platform. During most of its history, the flat snare was enclosed in the original maze of crossed threads. At first the lines of the platform intersected haphazardly to form an irregular framework made of dry dragline silk spun from the same glands as in modern forms. Over this skel- eton was laid a covering silk produced by different glands and dispensed through the posterior spinnerets, with which were mixed draglines from the ever active front spinnerets. These two elements, the framework and the covering, remain discrete throughout the evolution of the aerial flat snares. This definite pattern underlies the sheet web of the linyphiids, even though the finished sheet may not appear to be based on a definite plan. These weavers begin at the center of the dome, put THE AERIAL WEB SPINNERS 177 down straight lines an inch or two long, then cross them by over- spinning shorter lines. These principal framework lines need only to be lengthened, and they become the radii of the orb weaver, which likewise puts down its rays from the hub outward. The primitive radii were numerous, closely spaced, and probably fre- quently branched so that the interval between adjacent radii at the edge of the web was little greater than that near the center. (The silk spiders still use this device to produce their strong net webs.) All these framework lines were originally dry draglines, and remain dry in all spiders. The webbing that crossed the radii was at first dry or very slightly viscous, a condition reflecting both the pres- ence of only small amounts of sticky silk and the failure to concen- trate it in heavier drops. In the early orb weavers, the webbing over the dry framework corresponded to the viscid spiral of the higher orb weavers. The tremendous accomplishment that it represented was the formalizing of an irregular maze into a series of regular lines crossing the radii at nearly right angles. The first regular lines were probably series of curves that covered a sector of the whole, then larger loops oc- cupying half the circle, and finally complete spirals that produced the relatively symmetrical orb web. These lines may well have been long dry rods covered with a viscous coating. At this time the formal round platform, entirely enclosed within the maze of criss- crossed threads, was still only a platform on which insects dropped. By slow stages the accompanying mazes, especially the one above the platform, were lost, but only in the highest orb weavers are they gone completely— and even here their vestiges may still be seen in the tangle that leads to the retreat and the hidden egg sac. The gradual inclination of the orb, and the final near vertical posi- tion, were inevitable refinements. The evolution of the orb web progressed hand in hand with changes in the silk and in the spiders themselves. The silk glands gradually became a voluminous part of the abdominal contents, and were able to produce silks of differing properties. (In some modern forms there are more than six hundred separate glands producing five different kinds of silk.) Viscid silk was manufactured in larger quantities, and, when concentrated on the spiral lines, changed the round platform from a stopping net to an adhesive snare. As these early spinners developed, various groups branched off the main line to become sidetracked at different development levels. Some come down as probable replicas of early spiders, and their webs are sig- 178 AMERICAN SPIDERS nificant as indicating intermediate stages. The basilica spider, Al- lepeira, still entirely encloses its web within a maze of threads. The labyrinth spiders, Metepeira, largely preserve the lower maze as a tangle placed behind the orb, in which the spider rests. The spiders that inherited the tradition of the formal orb trap comprise a multitudinous group, of which many are familiar because of large size, bizarre form, and bright coloration. They are specially noticeable during the fall months, when their orbs, and the spiders themselves, attain maximum size and cover the vegetation in great profusion. Many of the spinners are fat little creatures that hang serenely in the hubs of their webs, head-downward, claws pulling the rays taut, poised to move in the direction of any disturbance. Others are less bold; they sit in the comparative security of a leafy nest, but they are attentive to the thread that communicates with the center of their snare. All are accomplished trapeze artists, and swing across the lines with grace and precision. They have pro- duced many different types of orb webs; but while their success must be largely attributed to the perfection of this trap, they have also sacrificed much to gain their pre-eminence among the space web spinners. They resemble the linyphiids rather completely in fundamental features. The cephalothorax is lower and wider in front; the eyes, invariably small and little used, lie near the front edge. The cheli- cerae are large and strong, and the maxillae are short and parallel, never pointed inward. The legs may be long and well spined, but they are frequently quite short and stout. They lack the stridulat- ing organs present on the chelicerae of the linyphiids. Sexual di- morphism is often very pronounced. In many instances, the males are quite safe within the bounds of the female's web, but not infre- quently she is an ogress. The developmental history of the orb web is only vaguely indi- cated in the spinning of modern orb weavers, which retain the essential details as an instinctive racial memory. The baby spider weaves its remarkably symmetrical web soon after leaving the egg sac, and thereafter, throughout its lifetime, modifies the plan only in minor ways. The spinning of an orb web is an involved process consisting of a series of separate steps. The spider must first delimit the area of operations by framing it with silken lines. The first and most important line is a more or less horizontal bridge on which the whole web is hung. There are two way of establishing this bridge line. A thread may be emitted from the spinnerets and floated in the THE AERIAL WEB SPINNERS 179 air until it catches on some object; whereupon it is pulled taut. Al- ternatively, the spider may fix a line, carry it, by dropping or walk- ing, down one side of the area to be covered, across and up the other side to the point of attachment, holding the line free of en- tanglement all the time. Once the bridge line is fixed, it is strength- ened by additional threads as the spider moves back and forth across it. (See Text Fig. 6.) From some point on the bridge line the spider now drops down to a lower point and fixes a plumb line to grasses, twigs, or any substratum. To this plumb line the spider next attaches a third line, and, holding it free, climbs back up to the bridge, along it for a distance, then attaches and tightens the third thread. There is now formed a triangle, its dimensions dependent upon the distance trav- eled on the bridge line, whose apex points down and in which the round snare can be placed. These foundation lines may be roughly rectangular, trapezoidal, or otherwise configured, and are dependent upon the local condi- tions and the habits of the species. Within the framework the spider now lays down the radii. First it must put down a diameter line to pass through the point that is to be the center of the orb. This may be accomplished by dropping down from the bridge line, or by walking the diameter line, held free, around the framework to the opposite point, where it tightens and fixes the line. From some point on this initial diame- ter line originate all the other radii of the web, each put down by carrying it around to the desired attachment point over the already fixed lines. According to McCook, the radii are as a rule laid alter- nately on the opposite sides of the enclosed space, but there may be less regularity that he supposed. The tension on the lines, no doubt susceptible to testing by the spider, may well influence it in setting down the radii, which are often placed with quite remarkable ac- curacy to form nearly equal angles at the center, but in other cases are grossly asymmetrical. After fixing the last radius, the spider usually goes to the point where the radii converge and strengthens it by spinning a mesh of lines termed the hub. A part of the hub is made while the radii are being placed and stretched, and it is completed after the last one is fixed. Around the hub are then spun several spiral turns, which, because they are laid down and pulled to form irregular notches, are termed the "notched zone"; they serve to strengthen the central area and tighten the radii. The next step is to put down across the i8o AMERICAN SPIDERS whole series of radii a spiral thread that holds them in place during the subsequent spinning. The turns of this "scaffolding spiral" are wide apart. Up to this point all the lines (the foundation, radii, hub, and scaffolding spiral) are of dry silk. They are the framework of the primitive platform. Beginning at the outer margin beyond the scaffolding spiral, the spider now puts down viscid spirals upon the dry web skeleton. It is guided, to some extent at least, by the scaffold, by tensions in the lines, by the nearness of outside lines, and by various other factors. Before the full circular turns begin, the spinning may be directed to filling in corners with short curves connecting few radii, or with longer loops that sometimes swing halfway around. As the spider gradually approaches the center, the dry lines of the scaffolding spiral are bitten out, wound up, and discarded, or sometimes eaten. The viscid spirals and loops, ordinarily composed of a single coiled thread from the beginning point to the end near the center of the orb, are placed down with a slow and deliberate motion in a very special manner. The spirals may be spun clockwise or counter- clockwise. These sticky lines are composite; they consist of a dry core of two closely joined threads covered evenly on the outside with a viscous film derived from different glands. Following attachment of the compound line to a radius, it is grasped by the claws of one hind leg; these, as the spider's body swings across the space, pre- pare to fix the thread on the next radius. At the same time, the line is grasped near the middle by the claw of the other hind leg, stretched rapidly to half again its length, then let go with a snap. The result is an elastic line that contracts to the width of the space between the radii; upon magnification, it is shown to be beaded with a series of small round drops of sticky silk. The stretching of the silk line breaks up the viscid film and distributes it along the line like beads on a necklace. The web is now essentially finished, and the spider returns to the hub, often to alter it in the way characteristic of its group, which consists of biting out the center or ornamenting it in various ways by adding distinctive bands of silk. These bands, decorative loops, or zigzags of thick, white, flossy texture, presumably strengthen the center of the orb; from this attribute they have received the name "stabilmenta." They seem to be vestiges of the early custom of overspinning the central portion to provide a resting space in PLATE XXIII •*"»'' "" ' f *' ';""' * " ' MUD DAUBER, MUD NEST AND SPIDER PREY Lynwood Chace Lee Passmore Gtorge Elwood Jenks a. Portrait b. The pendent egg sac, opened to show young THE BOLAS SPIDER, Mastophora cornigera PLATE XXIV George Elwood Jenks Walker Van Riper Feather-foot spider, Asymmetrical orb web of banded Uloborus americanus, with egg sac Argiope, Argiope trifasciata JRS George Elwood Jenks Female of tuberculate Cyclosa, Cyclosa turbinata, on egg string TEXT FIG. 6.-SPINNING THE ORB WEB A. The foundation lines delimit the snare. B. Radii are laid down in the frame. C. A dry scaffolding thread spirals out from the hub. D. The viscid spiral coils are laid down from the outside and the scaffolding thread removed. THE AERIAL WEB SPINNERS 181 which may have been slung the egg sac. The habit is largely lost by modern types, but some of the weavers have revived and ex- tended it to produce striking patterns over considerable areas of the web surface. It may be added that these bands, along which the spider sometimes aligns its legs, doubtless give protection from certain enemies. Recapitulating, the finished orb web consists of various distinc- tive parts. A strong, moderately elastic framework, strengthened by overspinning each line one or more times, holds in proper tension a series of radial lines of dry silk. At the center is a thickened or meshed hub, with or without extensive bands of silk, being the strategic center of the web, the place where the spider hangs with its claws touching the radii. Beyond the hub is usually a free zone devoid of spiral lines; then comes the series of sticky spiral coils that act to snare prey. The spider hangs downward or away from the web, even when it is nearly vertical, and moves by grasping the dry lines with tiny claws. The spider is anointed with an oil that to some degree prevents the sticky lines from adhering to it. When an insect becomes entangled in the lines, the whole web is agitated by the struggles, and the vibrations are communicated to the spider. Quickly it swings across the web to the site of the dis- turbance, directed by the pulls on the lines, all the while trailing behind it a dragline thread, on which it can drop to the ground or save a fall if brushed from the snare. Its long legs tap the prey, further informing the spider of the nature of the victim, and bring- ing on a response commensurate with the problem of subduing it. Small, weak insects are seized and quickly enwrapped, but larger and more active ones are treated with greater caution. The prey is seized, held by some of the legs and turned round and round, while it is trussed up with silk. Jets of fine filaments, thrown out al- ternately by the spider's hind legs, are combed over the insect and envelop it like a mummy. The bite may be administered either during the capture or later, when the prey is caried back to the hub and feeding begins. The struggling victim often cuts and entangles many spiral seg- ments and radii, and may cause whole sectors to sag. During the capture, broken lines are tied together with threads by the spider, which deftly grasps the lines with its claws and pulls them together around the rent. This effective repair of the web is carried out at other times as well, by some species; still others allow the snare to become a shambles. i8z AMERICAN SPIDERS Mention of web repair always brings to mind the views of Fabre and others who regard spiders as creatures that, while spinning their webs, must run through the same series of inflexible and in- stinctive actions from start to finish. The spider can build another orb with the pattern and peculiarities of its clan, but cannot repeat an earlier step out of its turn, cannot repair a rent in the lines. By cutting some of the lines it can be shown that the mechanical spider often spins blindly to produce an imperfect caricature of a snare. However, spiders seem not to be quite the automatons that this view demands. Many higher orb weavers spin a new web almost every night. Retaining the foundation lines intact, they remove the ragged threads and law down fresh radii and capturing spirals. (It is at this time that they frequently eat the rolled up balls of silk.) Their webs are prepared for a single or for only a few captures; conse- quently there is little incentive to repair a web that will not be used again. Even so, much informal repair, by overspinning broken areas, does go on. Some of the orb weavers use the snare for a longer period, and probably do far more informal repair of the rents. It is well known that the silk spiders replace only a part, usually one half, of their large web at one time. In other instances, there may be quite formal repair by replacement of a series of loops, or by keeping intact areas of loops and adding only the spiral or complete turns. The seem- ingly tremendous task of completely replacing an orb web ordi- narily requires less than an hour. The spinning is often done during twilight, when it is easy for an observer to watch the whole process, but others spin during the early morning. The principal groups of orb weavers are well represented in the temperate region of North America. Much of what we know about their natural history must be credited to the energy of the Reverend Henry C. McCook, whose fascinating and comprehensive three- volume work, American Spiders and Their Spinnings or k, is one of the classics in arachnology. Superbly illustrated and still authori- tative, this is a primer to which layman and spider specialist alike can refer for dramatic essays on our orb-weaving spiders. That all orb weavers represent a single series is undoubted, but that they should all be placed within the single family Argiopidae, the most common practice, is debatable. Some modern types repre- sent lines early detached from the main stream, and now forming PLATE 25 a. Wolf spider, Geolycosa turricola, side view J. M. Hotlister (^ m >. Burrow of wolf spider, Geolycosa, in grass WOLF SPIDERS J. M. Hollister PLATE 26 Walker Van Riper, Colorado Museum of Natural History Grass spider, Agelenopsis, on egg sac THE AERIAL WEB SPINNERS 183 isolated groups only imperfectly bridged by intermediate forms, if at all. The big- jawed spiders of the subfamily Tetragnathinae are in certain respects among the most generalized of all orb weavers. Most are greatly elongated spiders with very long, thin legs; the chelicerae are of great size, especially those of the males, which often project forward in a horizontal position. During mating, the chelicerae of the female are gripped in those of the male by means of long spurs that clamp the fangs, and, thus firmly hooked, are rendered impotent. Most tetragnathids live in grassy areas, and are especially common on the border of swamps and along streams. Some place their snares horizontally over and close to the water, but more frequently they are inclined or vertical, and framed in grass or shrubs. The snare, which has an open hub, is quite delicate, and is best suited to the capture of midges, mosquitoes, crane flies, and other small insects with weak flight. The stilt spiders of the genus Tetragnatha, which appress their slender bodies and legs closely against stems or hang as inanimate straws in the center of their webs, are the best-known members of this series. When disturbed, they drop on their draglines, often to the surface of the water, over which they stride like aquatic bugs. A dozen or more species occur in our fauna, most of them widely distributed and abundant. One of the largest is the half-inch-long Tetragnatba elongata, a grayish stick spider with great jaws longer than its carapace. Even commoner is Tetragnatha laboriosa, a smaller, yellowish species with a silvery abdomen, which lives in grass, often in dry areas. The thick-jawed spiders of the genus Pachygnatha resemble the stilt spiders, but their chelicerae are shorter and heavier, inclined downward, and their legs are shorter. They live near the soil in deep grass or under debris in damp places, cattail swamps being especially favored. They do not spin a usable web, but wander in search of small insect prey, as do the short-sighted vagrant spiders. They are limited quite largely to the north temperate zone, and are replaced southward by smaller, globose species that still use an orb web as a means of capturing insects. These latter belong in the genus Glenognatha, and differ in having the single tracheal spiracle advanced far in front of the spinnerets. Our best-known species is Glenognatha foxi, an eighth-inch-long, pink and silver spider of quite globose shape, which lives in meadows and grassy situations all over the South. Its delicate orb web, three or four inches across, 184 AMERICAN SPIDERS is usually found anchored in a horizontal position about two inches above the ground, tied to grasses and weeds. A much larger species, Glenognatha emertoni, one fourth of an inch long, lives in the Santa Rita Mountains and other high ranges in southern Arizona. The members of the subfamily Metinae are closely related to the tetragnathids, but are more advanced in their structural features and far more diversified in their general appearance. All of them spin the orb snare, leaving the hub open; some of their webs are remarkable atypical creations. The species of Leucauge are bril- liantly colored spiders, green and silvery white, often spotted with gold, orange, or copper; and our commonest species, venusta, well merits its name. The species of Meta, whose best-known member is the half-inch-long, brown and yellow Meta menardi (common in Europe and possibly introduced from there into this country), approach the typical orb weavers in form. Their inclined webs are placed in dark places, often in shallow caves, and the large, snow- white egg sacs are suspended by a short thread from near-by walls. The most interesting member of the Metinae is the basilica spider, Allepeira conferta, a moderately elongate creature one third of an inch long, whose cylindrical abdomen is furnished with a hump on each side near the base. It much resembles Leucauge. Its web is of especial interest, since it is to a large extent intermediate between the sheet web of the linyphiids and the typical orb web. Often placed deep in well-shaded spaces in bushes, this snare con- sists of a large maze of intersecting lines that include a light sheet web and additional irregular lines. The dome is an orb web, con- structed of a large number of closely spaced radii and crossed by a spiral line of presumably viscid silk, that has been pulled and guyed into dome shape. (See Text Fig. 5, D.) Largest of all orb weavers are the silk spiders of the subfamily Nephilinae, exotic dwellers of the tropics, whose bodies are often more than two inches long and whose thin legs sometimes span eight inches. Their great round webs of golden silk, which will run over three feet in diameter and are supported by lines of great length, are found spanning forest paths or hanging high in trees. The giant female is attended by pygmy males scarcely longer than her cephalothorax, which, although almost too small to be accept- able as food, must still cautiously approach her only after prelimi- nary tweaking of the web threads. People who walk along paths in deeply forested areas frequently stride into the tough lines before they see them. Small birds are easily ensnared, and quickly make THE AERIAL WEB SPINNERS 185 their plight hopeless by their struggles, which bind the many lines together into strong bands. The use to which the silk of this spider has been put by primitive peoples and its failure as commercial silk have been noted. Our only silk spider is Nephila clavipes (Plate 10), a long- bodied, olive-brown species with lighter spots on the abdomen and long legs provided with thick brushes of conspicuous black hairs. Now largely confined to the extreme southern states, this Nephila was probably the same one that lived much farther north at Floris- sant, Colorado, during Oligocene times. The bodies of the older females are fully an inch long, specimens from the tropics often far larger. The quarter-inch-long male is less than one percent of the female's bulk. The orb webs of Nephila are not replaced as frequently as those of other orb weavers, since they have features that make them far more permanent. The dry spiral scaffolding line is looped back and forth, and is a permanent part of the web. The radii are numer- ous; they are branched so that the interval at the outside of the web is little more than near the center. The viscid spirals are loops for the most part, only rarely complete circles. The hub is eccentric, and is located high up near the side of the web. The ray spiders of the family Theridiosomatinae are small, globose orb weavers, which have diverged rather sharply from the more typical members of the group, and seem to lie in the vague intermediate zone between the three principal families of aerial spiders. Our best-known species is the widely distributed Theridio- soma radiosa, once believed to be the same as a species found in Europe. Females run about one tenth of an inch long, and have rounded, oval, highly arched abdomens marked with many small silvery spots. The remarkable egg sac, a brownish, pear-shaped bag, is suspended by a long, often forked thread from the branches of trees or the sides of stones. The aerial station makes it immune to depredation by crawling insects. The ray spider is commonest in dark, damp situations, and favors shaded woods, the underbrush along streams, and nooks at the base of cliffs. The web, usually vertical in position and three to five inches in diameter, is a most remarkable structure. It is first spun as a reasonably typical orb snare with a meshed hub and sev- eral spiral turns in the notched zone, then the hub and these threads are bitten out. The radii, a dozen or so in number, are next rear- ranged so that they converge upon a small number of lines that 186 AMERICAN SPIDERS radiate from a point at or near the center. These rays, in turn, con- verge upon a short trap line that is attached to a convenient twig or surface. The spider rests its body on the rays or the orb, and, sitting upright and facing away from the snare while holding the slack line loosely between its front legs, pulls the web into the shape of a cone or funnel— or an umbrella turned inside out, as McCook described it. When an insect strikes the web, the spider lets go the line; and both snare and spider spring back to aid in the further entanglement of the victim. Only one of the rays, com- prising three or four radii, is badly damaged with each capture; so the spider uses the trap several times. The upright position is most unusual for an aerial spider. It is made possible by the spider sitting upon a foot basket of taut lines and clinging with its hind legs. The resemblance of this device to that of the triangle spider is most striking. Hyptiotes uses a single sector, hangs back-downward from the ray threads, and springs forward on the line when the trap is released. A single capture destroys the triangular snare, but Theridiosoma has several sectors in reserve. In the subfamily Argiopinae are handsome orb weavers second only to the silk spiders in size, and far better known, especially in the North, where their bright colors and large webs make them conspicuous creatures. More closely allied to the typical orb weavers than to the groups already mentioned, the Argiopinae differ from these former in having the posterior eye row strongly curved backward. The typical web of Argiope, the principal genus and the only one that will be discussed here, is ordinarily somewhat in- clined, but may be nearly vertical. It is provided with a sheeted hub. Frequently the web is accompanied by a tangle of lines be- hind the orb, the so-called "barrier web," and less occasionally by a thinner tangle in front. These are probably vestiges of the stopping mazes of the primitive orb weavers; as was the case with these pro- totypes, they provide a protective screen against some enemies. At the center of the web is a stabilmentum consisting of a zigzag band of white silk in nearly vertical position, usually occupying a third of the diameter of the orb. In some instances it is vaguely indi- cated, but ordinarily it is a conspicuous mark, a signature of this group of spinners. The spider hangs at the hub, head-downward as usual, with the legs at each side of the stabilmentum; by appropriate stimulation it can be induced to shake the web vigorously (as do the long-legged cellar spiders and many orb weavers) until it be- comes an indistinct blur. The males are very much smaller, about THE AERIAL WEB SPINNERS 187 one fourth as long, and mature much earlier than do their mates. In midsummer they are often found in tiny, imperfect webs near the snares of immature females; later they lurk in the threads of the barrier webs of mature, greatly enlarged females. Three well-marked species, widely distributed in both North and South America, are almost the only members of this striking group that have penetrated into the New World from their head- quarters in the Oriental and Australian regions. The silver argiope, Argiope argent at a (Plate XVIII), is a comon and characteristic spider of the American tropics, and reaches to our southern states, where it is locally abundant from Florida to California. The center of its web is provided with a two-banded stabilmentum forming a cross of white silk; the spider, mostly metallic silver and yellow, with the abdomen divided behind into rounded lobes, lies with legs stretched out in pairs to cover this zigzag cross. Orange Argiope aurantia (Plates 2 and 21; Plate II), the females of which have bodies running to more than an inch in length, is mostly black; the abdomen, with a pair of low humps at the base, is marked above with bright yellow or orange spots. The legs of young females are conspicuously ringed in black and white, but in adults they are usually all black. The large webs, often two feet in diameter, are placed upon shrubs or herbaceous plants along roadsides, in gardens, and around houses, also in meadows and marshes. The spiders usually remain in the center of their webs even during the hottest and sunniest days. Many flying insects are captured in the snare, but a favorite food is grasshoppers, which abound on the web sites. The large, pear-shaped egg sacs (whose spinning has been described) can be seen tied to shrubs in the fall or early spring. The banded argiope, Argiope trifasciata (Plates 14, 19, and 20; Plates I and XXIV), rivals the previous species in abundance, espe- cially in the West, but not in size. The abdomen is evenly rounded, oval in shape, without humps, and usually silvery white or yellowish and crossed by narrow, darker lines. A very beautiful spider, trifasciata, lives in essentially the same locations as Argiope aurantia, but may often be found in drier situations. The snares are very similar, and placed to entrap the same kinds of insects. The egg sac is cup-shaped, with a flattened top. The spiny-bodied spiders of the subfamily Gasteracanthinae are brightly colored creatures whose hard, leathery abdomens are orna- mented with prominent spines. The spinnerets are located at the tip of a conspicuous elevation. Several of these spiders occur in the i88 AMERICAN SPIDERS United States, but most live in the tropics, where an amazing array of bizarre types has developed. The genus Gasteracantha is poorly represented in the Americas; only three or four extremely variable species are found. The genus Micrathena, comprising more than a hundred long-bodied species, with flat or elevated abdomens bor- dered by long spines or thick spurs, is exclusively American. The spiny-bodied spiders hang on short legs in the centers of their webs, looking rather like chips of wood, bits of leaf, or plant fruits. The sharp spines make them unpleasant morsels for birds, lizards, and vertebrate animals, but their worst enemies, the solitary wasps, fill mud cells with them, not at all deterred by the armor. G aster acanthus webs are inclined or vertical, and have open hubs. The radii or foundation lines are ornamented with a series of floc- culent tufts of whitish silk. It has been suggested that these may serve as lures for midge-eating insects, which, deceived by the white flecks, might fly into the orb and be caught. Two species of Gasteracantha are found in the United States, but one, tetracantha, is very rare. The other, Gasteracantha cancri- formis (Plate 14), is common in the southern states. It is subject to considerable variation. It has a yellowish or orange abdomen spotted with black and fringed by six spines. Our species of Micrathena are equally spectacular in their bright coloration and curious shapes. Micrathena saghtata (Plate 22), an arrowshaped species having a white or yellow abdomen armed with a tiny basal and median pairs, at the apex a greatly enlarged pair, of divergent redtipped spines, is common even in the Northeast. Lumpy Micrathena gracilis, a light brown species, whose elevated abdomen is set with five pairs of short spines, is representative of a quite different series. These, and others not mentioned here, have small males that in shape do not closely resemble the females. The typical orb weavers of the subfamily Araneinae so far out- number those of other subfamilies that in the temperate regions they are the dominant group. Their physical characteristics are a generally thick-set appearance, bulky abdomens and relatively short legs, but some have become elongate types that can run quite rapidly. The abdomens of the typical orb weavers are subject to a very considerable variation in shape. Some are leathery, and sur- mounted by humps or spines that make them resemble those of the spiny-bodied spiders, from which they differ in not having the spin- nerets on the end of a tubular eminence. Sexual dimorphism is pronounced in many genera, especially in the bolas spiders; in cer- tain cases the males may be essentially equal in size to the females, THE AERIAL WEB SPINNERS 189 except for the bulk of the abdomen. It is among these orb weavers that the male seems to run the greatest risk while courting, but he has learned to reduce the danger by dropping on his dragline when his attentions are evidently unwelcome. His long front legs are usually armed with rows of spines that aid in holding the female, and in keeping her at arm's length when she pursues him. With the exception of the few species that have modified their ensnaring habits, the typical orb weavers spin a circular web, but considerable differences are found in the details of the orb and accompanying retreats. The small, elongated species of Cyclosa (Plate XXIV), usually about one fourth of an inch long, have conical humps at the end of the abdomen. Through the center of their beautiful snares, fur- nished with many radii and closely set spirals, often lies a stabilmen- tum consisting largely of the remains of insects and debris tied together with silk. The eggs are later added to the string. The spider sits at the hub, bridging the space between the segments of the string and blending so completely with it as to be practically invisible. The webs of most orb weavers still maintain, at least in vestigial form, the ancient mazes of their prototypes. In the labyrinth spider, Metepeira (Plate 23 and Text Fig. 5, E), the maze has been retained as a prominent, irregular net, which the spider uses as a base and in which it hangs its leaf retreat and string of eggs. Many species occur in the United States; they vary in size from one-fifth to one-half inch in length. The orb web is usually complete, with several trap lines leading from the hub to the retreat of the spider in the tangle. The spiderlings use the labyrinth as a nursery web after they break out of the egg sacs, and it is reported that they sometimes feed upon small insects caught in the tangle. Several of these typical argiopids habitually spin incomplete orb webs, entirely omitting the spiral lines and radii from a segment equal to the space between two or three radii. This they accomplish by spinning rounded loops, and swinging back and forth many times instead of making complete circles. Associated with such snares is a trap line, sometimes virtually bisecting the open sector but usually in a different plane, that leads to a nearby retreat. The species of Zygiella spin incomplete orbs of this type, but better known to Americans are those of Aranea pegnia and A. thaddeus, which have far wider distribution. The lattice spider, Aranea thaddeus, is about one-fourth inch long and has a rounded abdomen of a pale, yellow- ish color with darker sides. Its beautiful silken retreat is usually i9o AMERICAN SPIDERS attached to a near-by leaf; inside the retreat lies the spider, hold- ing the trap line stretched to the center of the orb. Spiders that use these trap lines first swing to the center of the web, then directly toward the point where the prey is entangled. A great many of our largest orb web weavers spin a complete orb and communicate with it by means of a trap line stretched from a retreat of folded leaves. Typical is the common shamrock spider, Aranea trifolium (Plates 7 and 18). This animal, often more than a half inch long, has an evenly rounded, white to pink abdomen, usually marked above with a three-lobed spot resembling a sham- rock. Its carapace is banded with brown, and its legs are conspicu- ously ringed with white and brown. The male is only about one fifth of an inch long. Even more stikingly colored is the related round-shouldered weaver, Aranea raji, which, common over most of the United States and well known in Europe, has a bright orange body and an abdomen marked by contrasting darker lines of an indistinct folium. Larger even than the round-shouldered orb weavers are some that have a bulky abdomen produced into prominent basal humps. These gray or brown spiders sit at the side of their complete orb webs in a crevice, under chips of wood, under bark, or in a more formal leafy retreat. One of the most familiar is Aranea gemmoides (Plate XXII), which is widely distributed from Nebraska westward. It varies from pale yellow to near black. The commonest eastern representative is Aranea nordmanni, a somewhat smaller and darker spider, with folium on the abdomen, which is thought to be similar to a European species. Included among the typical orb weavers is one small group that has repudiated the orb web in favor of a distinctive and extraordi- nary method of capturing insects. These are the bolas spiders of the genus Mastophora (Plates III and XXIII), fat creatures of above- average size, whose bodies are ornamented in a most grotesque manner. The carapace is bedecked with sharp, branched crests or horns, and set with many small, rounded projections; the volumi- nous abdomen is lined and wrinkled and surmounted with rounded humps. These bizarre specializations, reminiscent of similar orna- mentation in the dinosaurs and other groups of animals, are not known to play an important part in the life of the spiders. The hunting site of female Mastophora is usually the outer branch of a shrub or tree, most often high off the ground. On this the spider hangs in plain sight. Mementoes of her previous activities are numerous silken lines that soon form a thin coating over the THE AERIAL WEB SPINNERS 191 twigs and the leaves. Hanging to the lines, or hidden among the near-by leaves, may be one or more egg sacs, beautifully and dur- ably made and featuring a long, coarse stem drawn of! from the globular base. During the daylight hours Mastophora clings to a twig or leaf, completely immobile, perhaps deriving some protec- tion from her resemblance to various inanimate objects. Even when handled, she shows only a momentary evidence of life, and may be rolled around in cupped hands like a marble. Few spiders are so completely inscrutable. But Mastophora is a creature of the evening and night, and as one watches her then in the performance of her marvelous routine, one forgives the earlier listlessness. The disappearance of the last rays of twilight is the signal for action; she takes up her position for the evening's sport. With plump body swinging from the ends of her legs, she moves to one end of a branch and affixes her thread to the lower side by pressing her spinnerets against the bark. Grasping this thread with one of her hind legs and holding it away from the branch, she crawls several inches farther along and pastes the other end firmly in place. The result is a loosely hung line, which she often strength- ens with an additional dragline thread. This strong trapeze is hung far enough below the branch to allow a clear space for casting. Moving to the center of the line, Mastophora now touches her spinnerets and pulls out, to a length of about two inches, a new thread that lies clear of the other. Keeping it attached to her spin- nerets and held taut, she combs out upon it quantities of viscid silk. Each hind leg alternates in producing the liquid, until a shin- ing globule as large as a small bead is formed. The spider now pulls this line out still farther, allowing the weighted portion to drop part of the distance to its natural point of equilibrium, then she turns and severs it just below the globule with the claws of one of her hind legs. The freed line swings back and forth like a pendulum, but the spider turns quickly and approaches it, searching and groping with her front legs until she is able to grasp it. Quickly she swings her massive body and seizes the trapeze line by the hind legs of one side, adjusting the casting line between her palpi and one of her long front legs. Poised and ready now is the boleadora, waiting— with the patience that characterizes her— for the approach of a suitable victim. (See Text Fig. 5, C.) Also aroused to activity at this time are many nocturnal insects, which soon fly along accustomed lanes, dipping down close to the foliage and fluttering in and out among the branches. A large- 192 AMERICAN SPIDERS bodied moth, its wings spreading nearly two inches, it great eyes shining red in the last rays of reflected light, dips down toward the hunting grounds of the spider. As the insect approaches, Masto- phora gives every evidence of knowing that a prospective victim is near. She moves her body and adjusts her line, as if in tense ex- pectancy. At just the right moment, when the moth comes within the reach of the line, the spider swings it rapidly forward in the direction of the flier. The viscid ball strikes on the underside of a fore wing, and brings the moth to an abrupt stop, tethered by an unyielding line, which will stretch a fifth its length before breaking. Fluttering furiously at the sticky end of the lasso, the moth makes every effort to free itself, but the spider is quickly on hand to give the final coup. She bites her victim on some part of its body. With the venomous bite resistance ends quickly; and the paralyzed moth can be rotated and trussed up like a mummy in sheets of silk. Mastophora then sets to work feeding on the body juices of her catch. This bountiful food supply will keep the spider busy for some time. After having satisfied her appetite, she cuts the shrunken remnant loose from the trapeze line and drops it to the ground below. Later in the night a second capture may be made, but Mastophortfs needs for food are usually well met by a single sizable victim. It must not be concluded that the life of this spider is quite as simple as the incident portrayed might indicate. Mastophora may wait in vain for a flying creature to come near enough for capture. In many instances, her aim may not be as accurate as pic- tured, or the prospective victim may be too large to be held even by the strong band of silk. But patience is one thing at which spiders excel, and Mastophora is no exception. Should no victim reward her after half an hour of waiting, she winds the globule and line into a ball and eats it. Quickly she spins another line, prepares another sticky bead, and resumes her vigil. How wonderfully complex is the pattern of instinctive activities that make up the casting habit of Mastophora! Although endowed with glands that produce silk in copious quantities, the spider bases her whole economy on a blob of sticky silk dangling at the end of a short line. And still not content with a niggardly use of this vital material, she eats the viscid globule if it is not put to use. The trapeze line, the pendulum thread, the viscid globule, and the in- stincts of a hungry spider, have in her combined to produce one of the most sensational of all devices for the capture of prey. CHAPTER X The Hunting Spiders T JL HE HUNTING SPIDERS ARE FOR the most part bold creatures that put only moderate reliance on silk to gain a livelihood and spend much of their life in the field. They run upright on the soil and on vegetation, and maintain this upright attitude even when on webs. Among them are conspicuous extroverts, whose open ways have earned them such names as "wolf" and "fisher" spiders, "running" and "jumping" spiders, and other names that describe quite suitably the characteristics of animals that pursue and overpower their prey by strength, speed, and alert- ness. Their strong, usually elongate and cylindrical bodies are pro- pelled by stout legs of moderate length, as befits runnmg creatures. Many are big-eyed hunters with keen sight that stalk their prey during the daytime. But at the same time we also find numerous allies of retiring, even secretive habits— short-sighted vagabonds that skulk under the dark security of debris, that come out only at night to grapple fiercely with small creatures touched by their groping legs. The line of the vagrants starts with the same shy, short-sighted ancestral spider that gave rise to the aerial sedentary types. Origi- nally far less venturesome than its cousins, this prototype retired to the cover of a stone or a crevice, where it deposited its eggs and enclosed them in silken sheets. Then around itself and the precious bag it spun a silken tube or cell, at first left open at both ends, later closed behind, or in front as well. In this compartive security it spent most of its time, departing only for short hunting forays, after which it dragged the prey back to be devoured at leisure. Allegiance to silk was a moderate one. Dragline threads were put down during the foraging; and the elementary subservience to these lines still remains firmly fixed in the habits of the boldest and swift- est of the vagabonds. Silk was used by the males for sperm webs, and invariably by the females to make the flattened egg sacs, which were composed of lower and upper sheets joined at the margins. 194 AMERICAN SPIDERS Few of the hunters have completely given up the silken cell as a base. Some found such comfort there that they have remained in it throughout their history, and have modified it only by embellish- ing the entrance with various types of webs. Ariadna lies in the tube and waits for insects to trip on the signal lines that radiate from the mouth. Many funnel-web spiders spin a little silken collar around the opening and await callers with like patience. The well- hidden funnel of the grass spiders provides a sanctuary from the gate of which the spider may survey its vast sheet web where drop leaping or flying insects. All these spiders are fundamentally hunt- ing types; they represent a very distinct line from those creatures that use the third claw as a hook to swing through space. The sedentary vagrants rarely produce aerial webs of consequence, and they emulate only poorly the superb devices of the aerial snarers. Most of the early hunting spiders found it advantageous to move away from the bonds of silk. Improvement in vision made possible a life of action far from the retreat even during the daytime, and some were quickly molded into swift vagrants, with little need for a fixed station. Two distinct lines have been followed by the higher hunting spiders: one culminates in the wolf and lynx spiders, and the other in the jumping spiders. Whereas the curved, unpaired claws were the prime determi- nants of the departure of the aerial spiders from the main line of spider evolution, these had little to do in laying down the path of the vagrants. The wolf spiders and their kin retain the median claw, but it is small in size and not used as a hook. No claw tufts or accessory claws are ever present, but in some of the heavy ground forms the lower surface and sides of the distal leg segments are covered with thick pads of hairs. The gradual development of better eyesight in their prototypes made possible longer and longer forays away from the cell, thus leaving the egg sac vulnerable to the attacks of predators. During the egg-laying season the females remained near their sacs to guard them from depredation, and fre- quently were on hand until the progeny had emerged and dispersed. The lynx spiders and funnel-web spiders still guard their eggs in this fashion. Other early spiders learned to mold the flattened egg sac into a round ball and carry it around with them by the mouth parts beneath the body. The fisher spiders still use this cumber- some method. Both the stationary vigil and the unwieldy ball put strong restraint on the normally active lives of these spiders. To ease the curb, some transferred the round sac to their spinnerets so THE HUNTING SPIDERS 195 that it could be dragged, a position that permitted normal hunting. This habit is the badge of the wolf spiders. In the remaining vagrant line, the unpaired claw has been lost and the tarsi are supplied with adhesive claw tufts that allow the spiders to climb with great ease. In some of the wandering ctenids— Cupiennius and its relatives— the fading median claw may still be seen beneath the claw tufts; it serves to bridge the gap between the three-clawed and two-clawed vagrants. The type reaches its acme in the big-eyed jumping spiders, which are the most alert and in many ways the most highly developed of all spiders. Another prin- cipal branch has been the series of laterigrade families culminating in the typical crab spiders. Finally, at the very base of the series are the six-eyed hunting spiders, the remnant of an ancient group that has retained many primitive features. THE WOLF SPIDERS The handsome wolf spiders of the family Lycosidae are expert hunters that have few peers among their kin, and among all araneids are excelled only by the jumping spiders. They occupy almost every variety of terrestrial habitat, and seem to be at home in all as dominant predators. Some are amphibious types that rarely stray far from water, skating over or diving under the surface when they are menaced. Others have become adapted for a secretive life in areas of shifting, open sands, into which they dig tunnels and on the surface of which they hunt during the night hours. Most nu- merous in prairie regions, the wolf spiders abound wherever a plentiful insect food supply is available among the grass roots, and where the sunshine penetrates all but the densest clumps. Many wolf spiders have deserted their hereditary silk-lined cell for a life in the sun. Others, more conservative, return periodically to the retreat; some pass much of their life there, leaving it only to hunt. Quite a few have improved the retreat by changing it into a deep tunnel in the soil, in certain instances closed by a movable trap door. Only one group of wolf spiders, Sosippus, has moved in the other direction— that is to say, toward a greater dependency on silk; it spins a sheet web similar to that of the grass spider. Except for mere size, which varies widely between tiny quarter- inch Piratas and giant Lycosas, an inch and a half or more long, there is a surprising similarity in appearance among the wolf spiders. 196 AMERICAN SPIDERS The elongate cephalothorax is usually high and narrowed in front, and bears eight eyes whose size and position immediately distinguish the lycosids from almost all other spiders. Set close together on the lower part of the face is a row of four small eyes that point for- ward and slightly to each side. Immediately above these are two very large eyes that point forward, and farther back on the dorsal part of the head are two large eyes that look upward. The spider is thus able to see in four directions, and, because of the size and acute vision of some of these batteries, can perceive moving animals at a distance of several inches. The legs and chelicerae are robust, as befits such powerful hunting creatures; the oval abdomen is of moderate size. The capture of prey by the wolf spider is marked by vigor and power. The spider pounces upon its victim and, holding the body in its strong front legs, bites and crushes with its stout chelicerae. The capabilities of this rapacious hunter are not without limitation, however, when contrasted with those of higher animals. Although keen and long-sighted among spiders, its vision hardly merits com- parison with that of many insects. Its prey is perceived by sight, but the character of the moving object is probably not at all evident until the spider touches it. The diurnal lycosids are undoubtedly able to make greater use of their eyes than the nocturnal types; but these latter are conditioned to respond to the slightest disturb- ance of the soil of their hunting ground. Furthermore, the wolf spiders have a tapetum that reflects light rays back through the eye retina, and presumably improves their night vision. The female wolf spider is, in the fashion of her sex, a creature of variable temper. Notorious for her rapacious activities, she nevertheless displays a solicitude for her eggs and young that can scarcely be matched by any other spider. The mother Pardosa, which it will be recalled encloses her eggs in a carefully molded spherical bag, attaches the sac to her spinnerets and drags it around with her (Plate XXVI) wherever she goes. It make no difference that it is often as large as she is; this egg bag is a precious thing to her; she will defend it with her very life, and will fight viciously to retain it. Her instincts are most powerful ones, but— ironically— she is easily fooled and will accept for a time, and almost without question, a substitute sac from which the eggs have ben pilfered, a piece of cork, or a wad of paper or cotton of the proper size and shape. After two or three weeks, her young develop to a point where THE HUNTING SPIDERS 197 they can leave their crowded quarters. The mother then bites open the sac at the seam, and within a few hours a whole brood of tiny spiderlings has climbed upon her back and huddled there in a mass (Plate XXV) . The cluster will completely cover her abdomen and much of her carapace, and very often is composed of more than one layer of spiderlings. It seems to be true that the mother must open the egg bag, and that without her assistance the babies will often perish. During the time of carrying the young, the mother engages in normal hunting activities, and her children must accommodate themselves to a strenuous life. She will run with great speed when pursuing or being pursued, turn to defend herself when cornered, and during all these wild gyrations the spiderlings cling to her back. When brushed off, they quickly crawl back upon their perch— if they have the opportunity. During this period they do not take food, a fact that has led to considerable speculation as to how they are able to survive. By some they were thought to derive energy from the sun and air. However, adult spiders are notorious for their ability to go without nourishment, and the spiderlings are equally tolerant. Their bodies are provided with a food supply, and this is adequate to maintain them until they start feeding. While they are riding on their mother's back, which may be for a full week, they are merely biding their time until the next molt, after which they will leave to take up separate lives in the grass, and will begin their own hunting activities. The spiderlings do drink water during their stay, and probably find a sufficient supply in the dewy film that often covers them fct night. They have been observed to move to water and take their fill when the mother stops to drink, then clamber back on her abdomen. The success of the wolf spiders in surviving is unquestionably due in part to the initial protection given the eggs and young by the mother; that is to say, by maintaining a vigil over the sac, by carrying it always with her, by seeing that it is broken open and that the young are permitted to emerge. Thereafter, however, the clustered spiderlings seem to remain with her through their in- clination rather than hers. She pays little attention to them, and abandons them if they fall off and cannot reach her of their own initiative. It is possible to give in this brief section only a glimpse into the lives of a few American wolf spiders. A wealthy fauna made up 198 AMERICAN SPIDERS of many distinct groups with fascinating activities awaits the en- thusiast who cares to investigate further. The wolf spiders of the genus Pardosa (Plate XXV) are small, but they make up in abundance what they lack in size. In physical appearance they feature large eyes occupying nearly the entire width of the head, which is quite precipitous on the side. More gracefully built than the typical lycosid, they have a slender body supported by long, thin legs set with long, black spines. Their slender tarsi lack for the most part the conspicuous brushes of the larger lycosids. Their colors tend to be dark, frequently black, but the cephalothorax is usually marked by a pale longitudinal stripe continuous with a light band on the abdomen. The heads and fore- legs of the somewhat smaller males are often brightly variegated with white and black patches of hairs, features believed to be dis- played during courtship activities. The Pardosae are true vagrants and do not use any retreat for long, wandering instead over the soil and low vegetation in moist areas. All are sun-loving creatures and abound in the spring, at which time the males become mature and cavort in front of the more plainly colored females. Except in the far north, where more time may be necessary for complete development, they live only one year; in the case of the males, months less. Noted for their ex- cessive agility, they climb into flowers and over plants, and the spiderlings are often seen ballooning in the fall. Dozens of species of Pardosa live in temperate North America and occupy many different habitats. The moss- and lichen-covered slopes of the Far North and the highest mountains support distinc- tive dark species. In the dried grasses of meadows and along road- sides live small species striped in black and gray. In the Southwest, rocks and bare sands along creeks serve as the homes for speckled species that are hardly visible when not in motion. Most Pardosae abound in damp, grassy situations near bodies of water. Many are amphibious, being able to run over the water freely and to crawl under the surface by holding on to plant stems. One of the British species, Pardosa purbeckensis, lives in the intertidal zone and takes to the water during high tides, in the manner described below by W. S. Bristowe: The following day was sunny and a lot of the spiders were actively running about, but as the tide rose, they retreated to the higher portions of the plants. Presently I saw one which I PLATE 27 Walker Van Riper, Colorado Museum of Natural History Crab spider, Misumena calycina, on flower Walker Van Riper, Colorado Museum of Natural History Crab spider, Xysticus gulosus, with prey THE HUNTING SPIDERS 199 had been watching touch the water several times, like a bather feeling the temperature with his toe before taking the plunge, and then it deliberately walked down the stem of the plant be- neath the surface, taking with it a bubble of air, caught by means of its hairy body. 1 watched several others, and the same thing occurred, and this is therefore, how they survive the high tide. I was puzzled at first by seeing that they dived long before they were forced to by the submergence of their plant, but this was explained by an individual that got dislodged, for it could not dive without the help of something firm to hold on to, and even the tip of a leaf swaying in the current was not sufficient aid. Although they can run over the surface, they are far more com- fortable beneath it, especially in the rough weather, so the wis- dom of their submerging whilst something firm remains to cling to becomes clear.27 Pardosa is a small lycosid. There are some smaller— the "pirates" of the swamp-loving genus Pirata, and the shy Trabea of shaded woods— but in the main the typical wolf spider is larger and more stoutly built, and will often attain notable dimensions. Most of the typical wolves belong to the genus Lycosa (from the Greek mean- ing "wolf," or "to tear like a wolf"; it is also the common name for the whole group), and strength is the keynote of their makeup. The carapace is low and the sides of the head broadly rounded, so that the eyes ordinarily do not occupy the whole top of the head, but sit in a group at the center of a dome. The rather short, heavy legs are often supplied with dense brushes of hairs beneath the tarsi and metatarsi. These typical wolf spiders (Plate XXVI) are very handsome creatures. Their bodies are evenly covered with a dense coat of black, brown, or gray hair, which gives them a velvety appearance. Paler markings of various kinds, arranged in spots, patches, and stripes, add variety to the rich coloration of the hairs. Whereas the upper part of the body tends to harmonize with the terrain, the underside of both body and legs is often boldly marked with black patches and stripes. Lycostfs egg sac is almost always white in color. The female molds it into a nearly spherical object, and, turning and spinning over the edges, leaves scarcely any evidence of the seam where the 27 W. S. Bristowe, "A British Semi-Marine Spider," Ann. & Mag. Nat.. History, (9), XII, pp. 154-5. 200 AMERICAN SPIDERS two sheets have been joined. It is in many ways a much more fin- ished piece of work than the flattened bag of Pardosa, and would appear to be better adapted for dragging. Many of these lycosids are very active day hunters. Various handsome and distinctively striped varieties abound in grassland and in grassy areas along roadsides over most of the United States. One of the best known is long-legged Lycosa rabida, whose gray cephalo- thorax is marked by two chocolate-brown stripes and whose ab- domen displays a median brown stripe margined in yellow. A close congener is punctulata, in which the dark dorsal stripes are conspic- uous, and the venter of the abdomen varied with a series of small black points and markings. The body of Lycosa hentzi, a Florida species, is yellowish, and resembles dried grass. All these striped wolves are good climbers and often ascend high into grass bunches and low shrubs. One of the most interesting habits of the Lycosidae is the ex- tended and highly developed tunneling practiced by certain species. The splendid burrows made by them were not, of course, perfected in a single step. We can trace their gradual evolution in the habits of their creators. At the outset the wolf spider took temporary refuge beneath a stone, and lined the area with its characteristic silken cell. But space requirements for the growing spider often made it necessary either to enlarge the cell or to abandon it. There- fore, in order to employ the first of these alternatives, the spider had to develop the use of its chelicerae as digging instruments, and of its silk to bind the soil so it could be removed from the premises. The primitive burrows that resulted from attempts to enlarge the living quarters were only shallow depressions in the earth immediately below the cell retreat— and many contemporary wolf spiders still dig this type of pit. But other species increased their proficiency, moved their burrows to favorable sites in the open, and dug tunnels. Some made a further improvement by erecting at the burrow's mouth an elevated turret to serve as a lookout. Developing along another line, a few lycosids have learned to cover the entrance with a movable lid similar to those of the trap-door spiders. All the burrowing wolf spiders of the United States are large spiders that live more than one year, and in some instances do not attain full maturity until the second year. The spiderlings establish their burrows soon after they leave the mother, and gradually en- large them as they grow. They dig with their chelicerae, which are not, however, provided with a rake as are those of the trap-door THE HUNTING SPIDERS 201 spiders. They tie the soil together with silk into little pellets, which they carry in their chelicerae and drop a short distance from the bur- row entrance. The walls of the vertical tunnel are lined with silk, a very important material in the construction of the domicile; and the spider's movements are facilitated by a ladder of webbing that allows it to climb quickly and surely to the surface. The considerable reliance of these wolf spiders on silk is further noted in the various refinements associated with the burrow opening; the turret, the win- ter and aestivating closures, and the trap door— all are dependent on it. The typical burrow (Plates 25 and 26; Plate XXV) conforms throughout most of its length to the size of its occupant, but an en- largement, usually in the middle portion, allows the spider to turn around and serves— in the exact sense of the word— as a living room. Because of cramped quarters, mating ordinarily takes place on the surface, after the males have enticed the females to come outside. As for their maternal habits, the burrowing wolves transport their egg sacs and young around with them, even while moving in and out of the narrow tunnel. They have learned to carry the sacs to the entrance, where they can be exposed to the rays of the sun; a mother will sit just inside the opening, and turn the bag over and over with her legs and palpi to warm all its surfaces. (This habit appears to be a necessity for nocturnal species, and for those that scarcely move outside the tunnel entrance during their day hunting.) Whereas the vagrant wolves are usually rid of their young a week or so after they have clustered on the mother's back, the spiderlings of the burrowers may remain with their parent for long periods, sometimes over winter in the tunnels. Collectors seeking specimens will find that the burrowing lyco- sids may occasionally be duped by using a decoy— an insect tied to a string, a wad of beeswax, or a stem to which the enraged spider will cling long enough to be pulled out of its burrow. When the spider sits near or has been coaxed to the entrance by some meth- od, a quick jab with a knife blade or heavy forceps will close the lower part of the tube and make capture easy. Digging the bur- rower out may prove a laborious undertaking if the tunnel is tor- tuous or established in rocky soil. Those that live in sand are easily taken with shovel or trowel, but it is a wise precaution to put a stem into the burrow, or fill it with dry sand, and then follow its course down. The spider will usually retreat to the narrow bottom 202 AMERICAN SPIDERS and lie there quietly, well hidden with soil and not easily discov- ered until completely unearthed. Along the margins of North American streams and in sandy fields live a number of pale species that may here be termed "bank wolves." The best known of these is Arctosa littoralis, a whitish spider one-half inch in length that is flecked with many dusky markings, and often blends remarkably with the sand or gravel on which it sits. It is quite at home in loose sand, and frequently digs a burrow in this material, binding the grains together with silk and encircling the entrance with a collar of small stones. Littoralis is widely distributed from Canada to southern Mexico. Most of the individuals seem not to dig any sort of burrow, and instead will be found hiding under stones along lake shores and water courses. The largest of our wolf spiders is Lycosa carolinensis, a mouse- gray spider that combines a vagrant life in the open with the more prosaic one of the burrow. Females of all ages can be found wan- dering about or hiding under debris, the adults often dragging their huge egg sacs or carrying their numerous young. In the north these inch-long creatures assume a uniform dark grayish- brown, and the whole venter of the body is jet black. Examples from Texas and northern Mexico are far larger in size, lighter in color, and have the venter speckled or banded with black. The burrows of carolinensis are most commonly encountered in open country on relatively dry hillsides and in prairies covered with a sparse growth of low plants. The upper part of the tunnel is always inclined, and the deeper part is often quite tortuous, lying among roots and stones. The entrance is large and may lack any external modification, though on occasion this great spider builds a high turret of grasses, sticks, or stones around the hole. A particularly interesting variant on the turret theme is that of Lycosa aspersa, the "tiger wolf." This handsome spider, dark brown in color and possessing stout legs marked by many pale yel- lowish stripes, lives in open woodland in our eastern states and digs its tunnel straight down six or seven inches into the rich humus. Around the mouth it erects a high parapet of moss and debris, and over the top of this spins a canopy, leaving an opening on one side only. On top of the canopy are placed bits of soil, moss, and leaves, so that the whole nest is well hidden and blends with its site. In many instances the canopy is more than just a rigid covering; it becomes a hinged lid that may be lifted and dropped to close the PLATE XXV a. A female Lycosa covered with young Walker Van Riper Edwin Way Teale b. Portrait of male Pardosa milvina L. W. Brownell c. Turret of burrow of Lycosa carolinensis WOLF SPIDERS PLATE XXVI a. With captured fly Lee Passmore Lee Passmort b. With attached egg sac WOLF SPIDERS, Lycosa THE HUNTING SPIDERS 203 opening, and in this form is comparable to the wafer doors of the true trap-door spiders. Mary Treat was the first to describe the burrow of the tiger wolf; she observed over an extended period the life and general activities of a colony of twenty-eight of these spiders. In spite of their well-camouflaged nests, half of which were sealed during most of August, all but five tiger wolves fell victims to the digger wasps during that month. Those that escaped had completely ce- mented down the lids of their nests until the wasp season was over. Such a tremendous toll seems to suggest that aspersa is no safer living underground than her several close relatives, which rarely dig into the soil, and then make only a shallow cavity. In the southeastern part of the United States live many large Lycosae to which the common name "sand wolves" may be applied with considerable accuracy. A representative species, the most widely distributed of the whole series, is Lycosa lenta, a pale wolf covered evenly with grayish hairs and only lightly marked above by a dusky pattern. Intensive daytime collecting in Florida, where these sand wolves are most numerous, rarely produces examples of the several different varieties; at night, however, under the rays of the headlamp barren areas and seemingly unproductive habitats be- come bejeweled with their eyes, and it is possible to capture quarts of specimens within a short time. These wolves are extremely abundant on white sands, where they lie quietly with their legs outspread. They have fine eyesight, but rely almost entirely on touch to capture insects. When the sand is tapped with a pair of forceps, the spider rushes over to grasp and wTrestle with the instrument almost as it would with normal prey. Most intriguing of all the sand wolf's reactions occurs when it is disturbed: It turns a somersault, dives into the sand, and disap- pears, leaving on the smooth surface no sign of where it has gone. Careful investigation shows that there is a well-hidden burrow closed by a perfectly concealed trap door. This door is coated above by a fine layer of sand; it is very thin, even thinner than the most tenuous wafer door of the trap-door spiders but essentially similar to it. The sand wolf opens the lid quicky and crawls head- first into the cavity, closing the door after her with her legs. The burrowing life has left such small imprint on the bodies of its practitioners that they appear to differ in no important respects from the vagrant wolf spiders. They produce subterraRean dwel- lings comparable in excellence to those of many trap-door spiders, 204 AMERICAN SPIDERS but without benefit of the specific modifications that the latter enjoy. Only in the "earth wolves" of the genus Geolycosa do we find features that suggest a first step toward true adaptation to a subterranean life. Many wolf burrowers have thick, round bodies and modified appendages, but in Geolycosa the cephalothorax is higher and more strongly arched than usual, and the chelicerae are unusually robust. The front legs are very stout and proportionately thicker in both sexes, and all the legs lack prominent dorsal spines. The earth wolves are confirmed exponents of a subsurface existence, and spend almost all their lives within the burrow. Extremely shy, they are reluctant to move very far from the opening even when capturing insects, and usually sit partially inside, ready to retreat at the slightest disturbance. Most other wolf spiders will wander a few feet from the opening to wait for prey, or even forage long and far from their tunnel retreat. These may be approached at night with a lamp and easily captured, but the nervous earth wolves must be dug out of the soil. The species of Geolycosa (Plates 25 and 26) are to be found over most of the United States and temperate Mexico. Some are yellow-brown spiders clothed with whitish hairs, but most have dark red and brown bodies, masked by a covering of slate-gray or brown hairs. The undersides of body and legs are usually marked with jet-black bands and spots. They dig their burrows from six to twelve inches into the ground— the depth being somewhat de- pendent on the character of the soil— and line the whole with silk. Ordinarily the tunnel goes almost straight down, and is enlarged in the middle portion or at the bottom. Some of the palest American earth wolves (such as