Comparative studies on the cocoon spinning habits of Osmia excavata Alf ken and Osmia pedicornis Cockerell (Hymenoptera, Megachilidae)

So far as the author\u27s investigation goes, in the nests of the so-called tube-renters of the Aculeate Hymenoptera, the very interesting and important fact is that the heads of the overwintering prepupae or the newly emerged adults are directed exclusively towards the entrance of the nest, so that the adult wasps or bees can safely fly out in the field from their own nests after the destruction of cell partitions by their mandibles when the flying season has come around. If one wasp or bee takes a reverse direction in the cell, it and the other individuals, if present, which are found in the inner cells in the same nest may lose their lives sooner or later, because the adult in question cannot break down the partition by the tip of the abdomen instead of the mandibles. By what factors do the wasps and bees direct their heads towards the entrance of the nest ? Concerning this problem, Yasumatsu (6) published in 1929 an interesting observation on the cocoon spinning habits of Trypoxylon obsonator Smith (Sphecidae) and pointed out that the most important factor concerned may be found on the structure of the partitions of the larval cells. The larval cells of Trypoxylon obsonator are divided by the mud-partitions, and the surface of each partition facing to the entrance of the nest is smooth and concave, while the opposite surface is rough and convex. Yasumatsu assumed that the full-grown larvae can distinguish this structural difference of the partition and spin their cocoons in normal positions or pupate directing their heads towards the entrance of the nest. The author conducted comprehensive observations and experiments on the cocoon spinning habits of Osmia excavata Alf ken and Osmia pedicornis Cockerell, and came to the conclusion that Yasumatsu\u27s hypothesis is, excepting the consideration on the effect of the pollen-honey mass, accepted in Osmia excavata and also, in some cases, in Osmia pedicornis. The detailed discussions of the results were given in this paper. The larval cells of Osmia excavata and Osmia pedicornis are divided by the mud-partitions. The surface of the partition facing to the entrance of the nest is smooth and concave, and the opposite one is rough and convex (Fig . 1) just as in Trypoxylon obsonator. One of the partitions of excavata is measured as 6 mm in diameter and 1.5 mm in depth at the centre, the radius of curvature being 3.75 mm. The size of the larval cells of excavata is usually smaller than that of pedicornis, namely, 6～8 mm in diameter and 8～12 mm in length in the former, and 7～11 mm in diameter and 12～15 mm in length in the latter. The experiments were conducted chiefly by using the cells of 6～7 mm in diameter and 10～12 mm in length in excavata, and of 7 mm in diameter and 15 mm in length in pedicornis. It is worthy of notice that the adult bee of both species can not change its direction in these limited or narrow(concerning the diameter) nests. Therefore, the coincidence of the direction of the apex of the cocoon with the concave surface of the mud-partition bears a very important biological significance. The cocoon of excavata (Fig. 2) is nearly ellipsoid in shape, having a small process apically, and is covered by thin outer layers. The author dissected 2,8605) cocoons of excavata and found the fact that the position of the head of the adult bee is on the side of the process of the cocoon (the apex of the cocoon) with only one exception. As the results of experiments and observations on excavata, it was revealed that (1) the rounded basal portion of the cocoon is always attached tightly to the concave surface of the partition (and a part of the body of the cocoon also to the wall of the cell), when a full grown larva is kept in an empty cell, even if the cell is kept in vertical or converted positions (cf. Fig. 5, A～A\u27, B～B\u27; Fig. 7, E～E\u27, F～F\u27), (2) the body of the cocoon tends to attach to the upper wall of the cell when a mass of excrements is present in the cell (cf. Fig. 5, C～C\u27; Fig. 6, B), and (3) the position of the cocoon in the cell has a close connection to that of pollen-honey mass (cf. Fig. 5, D～D\u27; Fig. 7, G～G\u27). As the author (1) stated in his former paper, the nesting behaviour of Osmia excavata is, after the selection of the tube, begun by the formation of the basal partition of the first larval cell, followed by the preservation of a pollen-honey mass which is adhered to the concave surface of the partition. Next an egg is deposited on it, and finally the cell is closed by the partition which serves as the basal partition of the second larval cell as well. Therefore, it is obvious that the security of emergence of the offsprings is guaranteed from the beginning of the nesting behaviour of the mother bees. On the other hand, the cocoon of pedicornis shows a strong antigravity character which has almost no bearings to the position and direction of the concave surface of the partition or to the position of the pollen-honey mass when the cell is held in vertical or converted positions (cf. Fig. 4, A～A\u27; Fig. 8, B～B\u27, C～C\u27). This character may fully be demonstrated in Fig. 9. It is of extreme interest that the cocoon of pedicornis is always suspended in a cell (Fig. 11), so that it may be named as \u22a cocoon of hammock-type\u22, thus making a contrast to \u22a cocoon of attaching-type\u22 of excavata. An interesting and significant difference between these types of cocoons is shown by the following experiments on the cocoon spinning abilities of both species. In a glass-tube of about 20 mm high kept in a vertical position, the cocoon of excavata is spun with or without difficulty when the diameter is less than 12 mm (Fig. 10, A and B), but the larva spins a peculiar cocoon which looks like a bird\u27s nest if the diameter is over 13 mm (Fig. 10, C). However, the cocoon of pedicornis is spun (and suspended) without difficulty in a glasstube of 15 mm wide and about 20 mm high (Fig. 11). Consequently, it may be assumed from these tests that a comparatively narrow cell is needed for the spinning of the cocoon of excavata, or in other words, a given base ?it must be the concave surface of the partition wall and a part of the wall of the cell in normal condition-----is necessary for the cocoon of excavata and a full-grown larva of pedicornis can manage to build up some bases to a certain degree for spinning its cocoon.安松はジガバチモドキの繭が常に巣の入口に向つてつくられることを観察し, それは虫室隔壁の面に意味があると考えた. 著者はシロオビツツハナバチとオオツツハナバチの営繭習性を検討し, シロオビツツハナバチの場合には安松の考えが一花粉団子の位置に関する考察を除いて一正しくあてはめられることを立証すると共に, ツツハナバチ類の繭には密着型とハンモック型の二型があることを見出し, その営繭性の相違を明らかにした. 即ち, 1. ツツハナバチの繭の一端には小突起があるが, この突起のある側に前踊, 蝋, 成虫の頭部が見出される事実から, 繭には前後の区別があると考えられる. 2. シロオビツツハナバチの繭は密着型であつて, 虫室を構成する隔壁の一方の側, 即ち滑かで円く凹んだ面(凹滑面)に繭の後端部を, 巣壁の一部に繭の胴の一部を密着させて営繭する. 従つてこの凹滑而の位置と向きに従つて繭の位置と向きとが決定されるが, この際, 凹滑面は営繭の一拠点として訣くべからざる要素である. 3. このようにシロオビツツハナバチの繭は凹滑面と巣壁の一部を拠点として営繭されるから, 重力との関係, 換言すれば繭を直立につくるという性質が極めて弱く, そのために巣が水平に保たれる場合は勿論のこと, 巣を乖直にして直立或は倒立に保つた場合でも繭は凹滑面に密着され, 常に巣の入口にむかつてつくられることとなる. 4. 更にシロオビツツハナバチの繭は花粉団子がおかれていた場所に営繭される性質が認められる. 花粉団子は凹滑面に接して貯えられるから, 摂食を終つた幼虫は営繭の足場をうるための容易さが与えられるように考えられる. 5. このようにしてみれば, シロオビツツハナバチの繭が常に巣の入口に向つてつくられるという事実, 換言すれば成虫の巣外への脱出を安全に保障しているという機構は, 成虫の営巣活動のはじめにその端を発していることが明らかである. 即ち, 成虫の営巣活動は, 巣筒の選定がおわれば, 先ず第1虫室基部隔壁の築造にはじまるが, このとき成虫は常に入念に仕上げを行うから, 成虫が接する側, 即ち巣の入口に向つた面は滑かで円く凹む. 隔壁が完成すればこれに接して花粉団子が貯えられるから, 摂食を終つた幼虫は与えられた凹滑面を拠点として営繭する. 従つて繭は常に巣の入口に向つてつくられることとなる. 6. シロオビツツハナバチではジガバチモドキのように繭内に排泄物をためこむ場所をつくらず, 幼虫の発育の問に, 相当量の糞が虫室内に排泄される. 繭は糞をさけて巣壁の上部に密着されるが, この事実は第一義的には営繭の足場の一部を巣壁の上部に求めたものと解釈され, 第二義的には間接的な排泄物の処理法と考えられないことはない. 7. シロオビツツハナバチの密着型の繭に対してオオツツハナバチではハンモック型の繭をつくる. ハンモック型の繭は営繭に際して隔壁面の性状との交渉の度合が密着型よりも低いものと思われるが,事実壁面の性状に左右されるよりは電力に反して直立の繭をつくる性質が強い. 従つてオオツツハナバチの巣を倒立した場合には繭は直立するために巣の入口とは逆の方向に向くこととなる. 8. オオツツハナバチの繭は内径15mm, 高さ20mmの空間にも容易に営繭されるが, シロオビツツハナバチでは内径12mm, 高さ20mmの空間で辛うじて営繭可能であり, 内径13～14mmではすでに営繭不能となる. これはオオツツハナバチの場合は繭がハンモック型であるために, 老熟幼虫は比較的自由に自ら営繭の足場をつくつてゆくことができるのに反し, シロオビツツハナバチの繭は密着型であるために最初に営繭の足場が与えられているか否かに営繭の成否がかかつていることによると思われるが, このようにハンモック型と密着型の繭には全く異なつた営繭習性が認められる. 9. シロオビツツハナバチの巣は内径10mm以下の所謂単式巣が普通であり, このうちでも内径6～7mmの巣が最も多く見受けられるが, このような内径の巣では虫室長は8～12mm程度で, それらの虫室(空間)が営繭に最も適したもののように考えられる. オオツツハナバチの巣はシロオビツツハナバチよりも大きく, 内径8～11mm, 虫室長は12～15mm程度で, このような内径をもつた巣内ではオオツツハナバチの成虫は自由に体のむぎをかえ得る場合が多く, 繭は巣の入口にむかつてつくられるとは限らないといわれるが――このことに関する観察は極めて不充分であるから, 将来詳細な観察が行われることが望ましい――内径7mm程度の巣で成虫が自由に体のむぎを変えることができないような虫室では, 繭は凹滑面の向きに応じて巣の入口に向つてつくられる事実が認められる