aphids. If the latter are too long neglected, they discharge the honeydew on the leaves, where the returning ants find and collect it, before they approach the insects by which it was voided. But if the ants visit the aphids assiduously, the latter seem to comply with their desires by hastening the moment of evacuation. This is indicated by the diameter of the exuded droplet; and at such times they do not eject the antmanna to a distance, but, so to speak, retain and hand it over to their attendants.

"It sometimes happens that the ants are so numerous on a particular plant that they exhaust the aphids with which it is covered. Under such circumstances they stroke the bodies of their nurses in vain and are compelled to wait till these have pumped up a fresh supply of sap from the stems. The aphids are by no means parsimonious, and if they have anything to give, never fail to respond to the ants' solicitations. I have repeatedly seen the same aphid yield several drops in succession to different ants that seemed very eager for the syrup."

FIG. 206.

Earthen aphid

Many aphids bear on the sides of the sixth abdominal segment a pair of tubules with terminal orifices, known as cornicules, siphons or nectaries. Réaumur and other earlier authors saw drops of liquid exuding from these as well as from the anal opening, and from such observations has come the erroneous statement that the honey-dew is a secretion of the siphons instead of being merely the excrement of the plant-lice. Linné, for example, says of these insects: "Pleræque duo cornua, postica abdominis gerunt, quibus excernent rorem melleum hæ formicarum vacca!" And although this error, which was also promulgated by Buckton in his well-known "Monograph of the British Aphids" (1881-1883), has been disproved by Witlaczil, Büsgen, Kolbe, Forel, Mordwilko and others, we still find it tenaciously retained in many popular works on ants, not to mention text-books of entomology. So careful an author as Comstock, in a book intended for students ("The Study of Insects," p. 157), says: "On the back of the sixth abdominal segment there is, in many species, a pair of tubes, through which a sweet, transparent fluid is excreted. . . . The fluid which is excreted through the abdominal tubercles is the substance known as honey-dew." McCook also perpetuates this old error by publishing

figures representing ants taking drops from the siphons of aphids. The discharge of the drops of honey-dew from the anus and not from the siphons, is so easily observed with a pocket lens or even with the naked eye wherever aphids abound, that there is little excuse for

FIG. 207.

The actual condi

repeating the blunders of Linné and his successors. tions are shown in the figure (Fig. 204) which, for lack of a better one, I reproduce from the recent papers of Mordwilko (1896, 1907).

What, then, is the function of the siphons? The early writers replied to this question with a variety of more or less fantastic suppositions, although in general they believed these organs to have a secretory function. Réaumur was the first to remark that the siphons secrete a liquid of a more yellowish color than that expelled from the anus, and subsequent authors have shown that the siphonal secretion. has a sticky and wax-like consistency. Büsgen finally pointed out the usefulness of this liquid. "I was able," he says, "to ascertain the function of the tubules by observing the operations of a lace-wing larva [Chrysopa], the so-called aphis-lion, among a drove of plant-lice. This pale yellow animal, with a dark dorsal stripe and a length of only a few millimeters, hatches from a small egg, which is found during the summer attached by means of a long stalk to the most various plants. The larvæ are usually found on the lower surface of the leaves, and are easily captured, notwithstanding their agility. Their food consists in great part of plant-lice which they seize with their sucking mandibles and hold fast till reduced to mere shrivelled bits of chitin. When these larvæ are to be used for experiment, they are best left to starve for a night and then placed among a drove of aphids suitable for observation. When attacking they strike their jaws from below into the body of the plant-louse with a sudden blow and at once begin to suck out its juices. When the attack happens to be rather awkward, the aphid has time to smear the secretion, which is at once discharged from the tubules, over the face and forceps of the larva, which is thus, at least temporarily, disconcerted and frightened. To be sure, I never saw a larva free an aphid which it had once succeeded in seizing. The secretion hardens on the larva immediately and thus forms a most uncomfortable coating, causing the creature to desist from the chase, while it cleanses its forceps and forehead. This consumes time and can only be accomplished by the aphis-lion's seizing some slender object, like the tooth of a leaf, for the purpose of rubbing off the secretion. While in such a helpless condition, it may itself fall a prey to enemies. which it need not fear at other times. Its agility and sucking jaws, capable of being opened to an extraordinary width, make it a formidable enemy even to the ants, which it may actually overcome in the dark. On two occasions after confining a brown garden ant [Lasius niger], the most frequent guest of our aphid droves, in a small box with an aphis-lion, I found the ant sucked dry in the course of a few hours.

In the day-time, however, I have seen the ants drive the larvæ

off their preserves. Still the above observations show that notwithstanding the protecting guard of ants, the aphids require a special weapon, which may annoy, though it cannot kill, their enemy.

"Even more striking is the use of the tubercles in the case of the lady-bird beetles (Coccinellidae), those well-known enemies of the aphids.

FIG. 208. Carton tent like the one shown in Fig. 207, but opened to show the

coccids and the walls of the edifice. (Original.)

On the approach of one of these beetles, the rose-aphis at first seeks safety in her long legs, which, when touched, apprise her of danger before the enemy can reach her body. Sometimes she merely turns to one side without drawing her proboscis out of the plant tissues, at other times she lets herself drop bodily to the ground. If change of place does not put her out of reach of danger she begins operations with her tubules and smears the whole forepart of the beetle with their secretion. Usually the volley does not fall so much on the beetle's face as on its prothorax under which the head may be so far withdrawn as to suggest that this retractility may be merely a means of protection in just such emergencies."

Mordwilko has shown that the siphons are best developed in certain species of Aphidide that live singly and not in droves or colonies and are not attended by ants, whereas these repugnatorial glands may be vestigial or completely lacking in the species habitually thus attended. This is certainly suggestive of their great importance as organs of defense. Many aphids, too, secrete a thick covering of white wax, which may protect their thin-skinned bodies from the attacks of some of their enemies. In this connection, Mordwilko has called attention to the long, projecting anal spine of certain aphids. This he interprets as a mean of defense against the ants themselves, since it must hinder them in directly imbibing the drops of honey-dew. In the same

way the Russian author interprets also the waxy secretion which may fall on and pollute the limpid drops of excrement, but this is certainly not true of all wax secreting aphids. The common alder blight (Pemphigus tessellatus) secretes wax in abundance, but is, nevertheless, eagerly attended by several species of ants. It is difficult to understand why some aphids should repel the advances of the ants when other species apparently derive so much advantage from their companionship, for, although no complete list has as yet been published of the species of aphids and the ants with which they are always or only temporarily associated, we know, nevertheless, that a few species of the former are definitely symbiotic with ants, and that there are others with more or less pronounced proclivities of the same kind. Wasmann (1894) cites Forda formicaria as regularly myrmecophilous and Shouteden (1902) mentions Paracletus cimiciformis as living only in the nests of ants. Both of these species are radicicolous, that is, they occur on the roots and not on the aërial portions of the plants. But there is also, at least in North America and Europe, a long series of radicicolous aphids that occur with more or less frequency in ant nests. Shouteden records no less than seventeen species, representing nine genera (Geoica, Forda, Tetraneura, Schizoneura, Pemphigus, Trama, Chaitophorus, Aphis and Microsiphum) as occurring in the nests of Lasius niger alone. In North America varieties of this ant, and especially L. nearcticus, brevicornis and the various members of the subgenus Acanthomyops, harbor in their subterranean galleries a great many aphids. Mrs. W. P. Cockerell (1903) has shown that these include species of Tychea and Forda (F. kingi, interjecti, lasii and pallidula). Mordwilko cites Lasius brunneus as living exclusively at the expense of species of Stomachis, which are found on the stems and leaves of plants. Further studies of ants and aphids by investigators familiar with the species of both of these groups will unquestionably bring to light many additional instances of such intimate symbiosis.

Much of what has above been said of the aphids will apply also to the scale-insects and mealy-bugs (Coccidae). These are even more sedentary than the aphids, and may also occur on both the roots and aëria! surfaces of the plants. They are, however, more largely confined to warm countries, whereas aphids are more abundant in temperate regions. Abdominal tubules are absent in coccids, but they protect their bodies by secreting a covering of powdery white wax or a hard or tough scale. Like the aphids, they excrete honey-dew, often in considerable quantities, from the anal orifice. The manna of Biblical tradition is now known to be the honey-dew of one of these insects (Gossyparia mannifera) which lives on the tamarisk. This excretion.