Physical properties of single phospholipid bilayers adsorbed to micro glass beads. A new vesicular model system studied by 2H-nuclear magnetic resonance.

Spherical supported vesicles (SSVs), a new model system consisting of single dimyristoyl phosphatidylcholine (DMPC) bilayers adsorbed to spherical glass beads with a narrow size distribution, were prepared at two different sizes (0.5 and 1.5 microns) and their physical properties were studied by deuterium nuclear magnetic resonance (2H-NMR). Such SSV samples can be prepared at any desired size between 0.3 and 10 microns. The 2H-NMR measurements provide evidence for a strong dependence of the spectra and the transverse relaxation times on the curvature of the SSVs in a diameter range between 0.5 and 1.5 microns. For larger SSVs (1.5 microns diameter) their powder spectra and their calculated oriented spectra are similar to those obtained for multilamellar dispersions of DMPC-d54. The lineshape of the smaller SSVs exhibits a temperature dependence which is not found in multilamellar samples. The SSVs are stable in the liquid crystalline phase over days but irreversibly change to multilamellar vesicles in the gel state. The average thickness of the water layer between the single bilayer and the glass bead surface was estimated by 1H-NMR to e 17 +/- 5 A.     

Fiber types and fiber diameters in canine respiratory muscles

In the present study, we measured fiber types and fiber diameters in canine respiratory muscles and examined regional variation within the diaphragm. Samples of eight diaphragm regions, internal intercostals, external intercostals, transversus abdominis, and triceps brachii were removed from eight adult mongrel dogs, frozen, and histochemically processed for standard fiber type and fiber diameter determinations. The respiratory muscles were composed of types I and IIa fibers; no IIb fibers were identified. Fiber composition differed between muscles (P less than 0.0001). Normal type I percent (+/- SE) were: diaphragm 46 +/- 2, external intercostal 85 +/- 6, internal intercostals 48 +/- 3, transversus abdominis 53 +/- 1, and triceps 33 +/- 7. The diaphragm also contained a type I subtype [6 +/- 1% (SE)] previously thought only to occur in developing muscle. Fiber composition varied between diaphragm regions (P less than 0.01). Most notably, left medial crus contained 64% type I fibers. Fiber size also varied systematically among muscles (P less than 0.025) and diaphragm regions (P less than 0.0005). External intercostal fiber diameter was largest (47-50 microns) and diaphragm was smallest (34 microns). Within diaphragm, crural fibers were larger than costal (P less than 0.05). We conclude that there are systematic differences in fiber composition and fiber diameter of the canine respiratory muscles.     

Probing the mechanism of incorporation of fluorescently labeled actin into stress fibers

The mechanism of actin incorporation into and association with stress fibers of 3T3 and WI38 fibroblasts was examined by fluorescent analog cytochemistry, fluorescence recovery after photobleaching (FRAP), image analysis, and immunoelectron microscopy. Microinjected, fluorescein-labeled actin (AF-actin) became associated with stress fibers as early as 5 min post-injection. There was no detectable cellular polarity in the association of AF-actin with pre-existing stress fibers relative to perinuclear or peripheral regions. The rate of incorporation was quantified by image analysis of images generated with a two-dimensional photon counting microchannel plate camera. After equilibration of up to 2 h post-injection, FRAP demonstrated that actin subunits exchanged rapidly between filaments in stress fibers and the surrounding cytoplasm. When co-injected with rhodamine-labeled bovine serum albumin as a control, only actin was detected in the phase-dense stress fibers. The control protein was excluded from fibers and any linear fluorescence of the control was demonstrated as a pathlength artifact. The incorporation of AF-actin into stress fibers was studied by immunoelectron microscopy using anti-fluorescein as the primary antibody and goat anti-rabbit IgG coupled to peroxidase as the secondary antibody. At 5 min post-injection, reaction product was localized periodically in some fibers with a periodicity of approximately 0.75 microns. In large diameter fibers at 5 min post-injection, the analog was seen first on the surface of fibers, with individual filaments resolvable within the core. In the same cell, thinner diameter fibers were labeled uniformly throughout the diameter. By 20 min post-injection, most fibers were uniformly labeled. We conclude that the rate of actin subunit exchange in vivo is extremely rapid with molecular incorporation into actin filaments of stress fibers occurring as early as a few minutes post-injection. Exchange appears to first occur in filaments along the surface of stress fibers and then into more central regions in a periodic manner. We suggest that the periodic localization of actin at very early time points is due to a local microheterogeneity in which microdomains of fast vs. slower incorporation result from the periodic localization of actin-binding protein, such as alpha-actinin, along the length of the fiber.     

Identification of alpha and gamma trigeminal motoneurons by the vibratome paraplast technique for HRP histochemistry

A morphometric analysis of the masseteric motoneuron pool of the trigeminal motor nucleus was performed in the rat using horseradish peroxidase as a marker. Thick (40 microns) cryosections and thin (7 microns) Paraplast sections were compared. Two types of motoneurons related to the masseter muscle were observed. Small motoneurons, which had a high nuclear index, were found interspersed between large motoneurons, which had more cytoplasm. Evidence is provided that the small trigeminal motoneurons are gamma neurons that innervate the intrafusal muscle fibers of the masseteric muscle spindles.     

Experimental measurements of the temperature variation along artery-vein pairs from 200 to 1000 microns diameter in rat hind limb

Theoretical studies have indicated that a significant fraction of all blood-tissue heat transfer occurs in artery-vein pairs whose arterial diameter varies between 200 and 1000 microns. In this study, we have developed a new in vivo technique in which it is possible to make the first direct measurements of the countercurrent thermal equilibration that occurs along thermally significant vessels of this size. Fine wire thermocouples were attached by superglue to the femoral arteries and veins and their subsequent branches in rats and the axial temperature variation in each vessel was measured under different physiological conditions. Unlike the blood vessels < 200 microns in diameter, where the blood rapidly equilibrates with the surrounding tissue, we found that the thermal equilibration length of blood vessels between 200 microns and 1000 microns in diameter is longer than or at least equivalent to the vessel length. It is shown that the axial arterial temperature decays from 44% to 76% of the total core-skin temperature difference along blood vessels of this size, and this decay depends strongly on the local blood perfusion rate and the vascular geometry. Our experimental measurements also showed that the SAV venous blood recaptured up to 41% of the total heat released from its countercurrent artery under normal conditions. The contribution of countercurrent heat exchange is significantly reduced in these larger thermally significant vessels for hyperemic conditions as predicted by previous theoretical analyses. Results from this study, when combined with previous analyses of vessel pairs less than 200 microns diameter, enable one estimate the arterial supply temperature and the correction coefficient in the modified perfusion source term developed by the authors.     

Effect of rate of change of luteinizing hormone concentration on in-vitro progesterone secretion within rat corpora lutea during differentiation.

Immature rats were injected with pregnant mares' serum gonadotrophin followed by human chorionic gonadotrophin (hCG). Ovaries were removed 0, 2, 5 or 8 days after hCG and either prepared for morphometric analysis or perifused with 0, 5 or 30 ng luteinizing hormone (LH)/min. In a second study, ovaries were removed on Day 2 or 8 and perifused with 0.1 mg 8-br-cyclic adenosine 5'-phosphate/ml (8-br-cAMP). On Day 0, the granulosa cells of the preovulatory follicles were small (53 +/- 0.5 microns2) with a cytoplasmic to nuclear (Cy:Nu) ratio less than or equal to 1.5. By Day 2, corpora lutea (CL) were present and composed of 95% small luteal cells (diameter less than 125 microns2, Cy:Nu greater than or equal to 3.0) and 5% large luteal cells (diameter greater than 125 microns2, Cy:Nu ratio greater than or equal to 3.0). The percentage of large luteal cells increased to 36 +/- 7% by Day 5, suggesting that they are derived from a select population of small luteal cells. Basal progesterone secretion increased from 38 +/- 5 on Day 0 to 1010 +/- 48 pg/mg/ml on Day 8. The rate of 5 ng LH/min stimulated progesterone secretion on Days 0, 2 and 8; 30 ng LH/min stimulated progesterone secretion on Days 0, 2 and 8, but not on Day 5; 8-br-cAMP stimulated progesterone secretion on both Days 2 and 8. These data demonstrate that once granulosa cells are induced to luteinize they lose their capacity to secrete progesterone in response to 5 ng LH/min and do not regain their responsiveness to LH rate until they completely differentiate. The loss of this LH responsiveness appears to be due to an inability to stimulate sufficient intracellular cAMP concentrations, since cAMP stimulates progesterone secretion on both Days 2 and 8.     

Nuclear magnetic resonance study of spin relaxation and magnetic field gradients in maple leaves.

1H Nuclear magnetic resonance techniques were used to measure the distributions of spin-spin relaxation times, T2, and of magnetic field gradients in both the chloroplast and nonchloroplast water compartments of maple leaves (Acer platanoides). Results showed that encounters between water molecules and membranes inside chloroplasts provide an inefficient relaxation mechanism; i.e., chloroplast membranes interact weakly with water molecules. Gradient measurements indirectly measured the sizes of chloroplasts by showing that water in the chloroplasts is confined to small compartments a few microns in diameter. A comparison between measured gradients and gradients calculated for a model leaf indicated that chloroplasts are somewhat more likely to occupy positions along cell walls adjacent to air spaces, but also they may be found in the interiors of cells.     

Implications of nuclear diameter in small cell lung carcinoma

The nuclear diameter of 5,117 malignant cells from 42 small cell lung carcinoma (SCLC) patients was assessed either on pretreatment tissue sections (35 cases) or cytologic smears (7 cases) by ocular micrometry. The SCLCs were subtyped as 30 oat cell carcinomas and 12 intermediate cell carcinomas according to the World Health Organization classification, based on the predominant histology of the tumor. The median number of nuclei measured from each patient was 110. All patients were treated identically by sequential hemibody and local irradiation combined with chemotherapy and had a median follow-up time of 310 days. The mean nuclear diameter (+/- standard error) obtained from tissue sections was 8.2 +/- 0.03 microns (median = 8.0), including 7.3 +/- 0.03 microns (median = 7.0) for oat cell cases and 9.5 +/- 0.06 microns (median = 9.0) for intermediate cell cases (P less than .001). In 28.6% of these patients, the nuclear diameter overlapped in the range of 8 microns to 9 microns between both subtypes. Comparisons between the nuclear diameter of primary and metastatic SCLC cells revealed no statistically significant differences. The nuclear diameter of malignant cells correlated with the mitotic index and stage of disease, but did not correlate with the other nuclear morphologic variables or with survival. The only identified prognostic factor was the stage of disease; these results indicate that the nuclear diameter of malignant cells should not be considered a prognosticator or a guide for therapy in SCLC patients.     

Changes in the distribution of sizes of ovine luteal cells during the estrous cycle

The ovine corpus luteum is composed of two types of steroidogenic cells, which are referred to as small and large luteal cells. In this study, the size and number of steroidogenic cells were determined in corpora lutea collected on Days 4, 8, 12, and 16 of the estrous cycle. Corpora lutea were dissociated into single-cell suspensions that were stained for 3 beta-hydroxysteroid dehydrogenase (3 beta-HSD) activity, a marker for steroidogenic cells. The size of 3 beta-HSD-positive cells was measured with a Zeiss Videoplan Image Analyzer. On Day 4, most of the 3 beta-HSD-positive cells were less than 18 microns in diameter, the median being 11.2 microns. By Day 8, the number of 3 beta-HSD-positive cells increased 3-fold, and the median diameter increased to 12.8 microns. Although the number of 3 beta-HSD-positive cells was reduced by approximately 50% on Day 16, the median size on Days 12 and 16 was 14.6 and 16.8 microns, respectively. The ratio of large (greater than 18 microns) to small (less than 18 microns) luteal cells was 0.11 +/- 0.03 on Day 4; the ratio increased linearly to 0.67 +/- 0.09 by Day 16. This increase between Days 4 and 12 was attributable to an overall increase in the size of the cells; the increase between Days 12 and 16, however, was due to a loss of small luteal cells. When the experiment was conducted near the end of the breeding season, before animals became anestrous, the median size of the luteal cells did not change at different times of the estrous cycle but remained constant throughout. These data suggest that development of the corpus luteum is associated with an increase in the size and number of steroidogenic luteal cells, and that luteolysis is associated with a preferential loss of small luteal cells.     

Pre-erythrocytic development and associated host responses to Haemoproteus meleagridis (Haemosporina: Haemoproteidae) in experimentally infected domestic turkeys

Two generations of pre-erythrocytic schizogony occurred in skeletal and cardiac muscle of domestic turkeys infected with sporozoites of Haemoproteus meleagridis. First generation schizonts reached maturity approximately five days post-inoculation (DPI) and developed in capillary endothelial cells and myofibroblasts. The schizonts ranged from 12 to 20 microns in diameter and produced long (5-6 microns), slender merozoites. Early second generation schizonts were first detected in capillary endothelial cells between 5 and 8 DPI. They were cylindrical and ranged in size from 5 to 8 microns in diameter and up to 28 microns in length. Second generation schizonts which reached maturity by 17 DPI were surrounded by a thick, hyaline wall and were packed with numerous spherical merozoites less than 1 micron in diameter. Mature megaloschizonts were fusiform, ranged from 30 to 113 microns in diameter, and extended as much as 465 microns along the long axis of muscle fibers. Merozoites developed as buds from cytomeres that formed between 8 and 14 DPI. Infected turkeys developed a moderate to severe myositis within 5 DPI and were lame in one or both legs. The myositis was associated with the necrosis of scattered groups of muscle fibers. Muscle fibers surrounding mature megaloschizonts were swollen and hyaline. Megaloschizonts were surrounded occasionally by fibroblasts and infiltrates of mononuclear cells. The morphology and site of development of mature megaloschizonts of Haemoproteus meleagridis are contrasted with those of other avian haemosporidians.     

Phosphorus-31 nuclear magnetic resonance study of post mortem catabolism and intracellular pH in intact excised rabbit muscle

Phosphorus-31 nuclear magnetic resonance has been used to study the post mortem catabolism of high-energy phosphate compounds and the associated intracellular pH variation in pure fast- and slow-twitch rabbit muscles and in rabbit muscle with mixed fiber types. Comparative results from pure fiber types are reported for the first time. Large amounts of glycerophosphorylcholine (14.1 mumol/g fresh tissue) are found in the internal conoidal bundle (ICB), a pure oxidative slow twitch muscle, whereas the m. psoas major (PM), a pure glycolytic fast twitch muscle and the m. gastrocnemius caput medialis (GCM), with mixed fiber types, are devoid of the same metabolite. The total content of phosphorylated metabolites is constant among the three muscle types. The time-dependent post mortem changes in phosphorylated metabolites display the expected rapid drop in phosphocreatine and a simultaneous increase in intracellular inorganic phosphate. However, the ATP level remains constant during more than 2 h. Rate constants for metabolite breakdown and apparent ATPase activity have been determined. The comparative kinetics of intracellular acidosis at 25 degrees C yield rates of 3.3 X 10(-3) pH unit/min for PM, 2.7 X 10(-3) pH unit/min for GCM and 3.0 X 10(-3) pH unit/min for ICB. Initial intracellular pH values are 7.07, 7.20 and 7.02, respectively. Upon aging, the heterogeneity of the Pi signal reflects the existence of cellular compartments with different internal pH. The results suggest that the more intense low-pH Pi signal arises from the sarcoplasmic reticulum while the less intense resonance would reflect the sarcoplasmic higher pH. The temperature effect on post mortem catabolism in the 15-25 degrees C range has been documented. As expected, phosphocreatine and ATP breakdown increase with temperature but at a higher rate for slow-twitch ICB than for fast-twitch PM.     

Extraocular muscles in the lamprey, Lampetra fluviatilis L. I. Muscle fibres

Six extraocular muscles of the river lamprey, Lampetra fluviatilis L., were studied with the light and electron microscope. On the basis of morphology and histochemistry three types of muscle fibres were distinguished: thin, thick mitochondria-rich and thick multifibrillar fibres. In the thin fibres, 2.8-22.4 microns in diameter, myofibrils are distributed peripherally and show strong ATPase activity. The mitochondria are located paraxially. In the thick mitochondria-rich fibres, 19.4-31.0 microns in diameter, myofibrils are also located peripherally, whereas the central part of the fibre is densely packed with very numerous mitochondria possessing tubular cristae. Thick multifibrillar fibres, with a diameter similar to that of the former type, contain thin myofibrils scattered over the entire cross-section of the fibre. The activity of myofibrillar ATPase is lower in both types of thick fibres than in the thin ones. The tubules of the T system were observed frequently only in the thick multifibrillar fibres. The extraocular muscles of the lamprey are composed of large quantities of muscle fibres. Thin and thick fibres do not form separate layers, but are more or less uniformly distributed throughout the muscle. Many muscle fibres show structural features suggesting their degeneration.     

High -resolution scanning electron-microscopic studies on the three-dimensional structure of mitochondria and sarcoplasmic reticulum in the different twitch muscle fibers of the frog

Summary The three-dimensional structure of the mitochondria and sarcoplasmic reticulum (SR) in the three types of twitch fibers, i.e., the red, white and intermediate skeletal muscle fibers, of the vastus lateralis muscle of the Japanese meadow frog (Rana nigromaculata nigromaculata Hallowell) was examined by high resolution scanning electron microscopy, after removal of the cytoplasmic matrices.The small red fibers have numerous mitochondrial columns of large diameter, while the large white fibers have a small number of mitochondrial columns of small diameter. In the medium-size intermediate fibers, the number and diameter of the mitochondrial columns are intermediate between those of the red and white fibers.In all three types of fibers, the terminal cisternae and transverse tubules form triads at the level of each Z-line. The thick terminal cisternae continue into much thinner flat intermediate cisternae, through a transitional part where a row of tiny indentations can be observed. Numerous slender longitudinal tubules originating from the intermediate cisternae, extend longitudinally or obliquely and form elongated oval networks of various sizes in front of the A-band, then fuse to form the H-band collar (fenestrated collar) around the myofibrils. On the surface of the H-band collar, small fenestrations as well as tiny hollows are seen. The three-dimensional structure of SR is basically the same in all three muscle fiber-types. However, the SR is sparse on the surface of mitochondria, so the mitochondria-rich red fiber has a smaller total volume of SR than the mitochondria-poor white fiber. The volume of SR of the intermediate fiber is intermediate between other the two.     

Distribution of different fiber types in whole cross-sections of human platysma. A histological and morphometric study

Analysis of cryostat cross-sections of the entire platysma muscle from human autopsies revealed enzyme histochemical and morphometric differences between normal human limb and facial muscles. The mean diameter of platysma fibers was about 50% of that of normal limb muscle fibers. Fiber type diameter increased from the medial to the lateral parts of the platysma. A variability coefficient of 356 indicated great variation in fiber caliber, with many fibers of 10 microns and less. Type I fibers showed an increase in density from the medial to the lateral parts of the muscle. The arrangement of histochemical fiber types was irregular with a tendency to form fiber type groupings.     

Radicalization of lignocellulosic fibers, related structural and morphological changes

The radicalization of unbleached lignocellulosic fibers obtained from thermomechanical (TMP) and chemothermomechanical (CTMP) pulps was performed in heterogeneous phase by reaction with dioxygen in the presence of N,N'-ethylenebis(salicylideneiminato)cobalt(II), [Co(salen)], as catalyst. Phenoxy cobalt radicals immobilized in fibers were observed by electron paramagnetic resonance (EPR) spectroscopy; their amount depends on the fiber swelling induced by reaction medium. The absolute concentration of such radicals in fibers, about 10(16) spin/g, reaches values 10 times higher than that of phenoxy radicals formed in similar oxidative reactions catalyzed by laccase. The generation of phenoxy cobalt radicals in fibers was related to structural changes of lignin units, detected by mono- and bidimensional nuclear magnetic resonance ((13)C NMR and 2D-HSQC) investigations, and to morphological modifications in fibers observed by scanning electron microscopy (SEM).     

Genetic control of mitosis. Premature beginning of telophase chromatin reorganization in cells of the Drosophila melanogaster strain with ff3 mutation

The effect of cell cycle mutation ff3 on chromosome segregation was studied on fixed cells of neural ganglia. The cell distributions by diameter of interphase nuclei and by distance between sister chromatid sets were compared at anaphase and telophase. In the control wild-type strain Lausenne, the cell distribution by distance between sister chromatids in anaphase was similar to their distribution by nuclear size. The mean distance between segregating chromatids at anaphase (lcp) coincided with the mean diameter of interphase nuclei (dcp) and was 8.3 microns. Cells passed to telophase when chromatids were at least 10 microns apart. The mutant ff3 strain differed from the control strain Lausenne in cell distribution by interphase nuclear diameter and distance between sister chromatids in anaphase; the mean nuclear diameter and mean distance between segregating chromatids similarly increased to 9.3 microns. A specific feature of mitosis in mutant strain ff3 was a premature beginning of telophase chromatin reorganization. This caused the occurrence of cells with abnormally short (less then the interphase nuclear diameter) distance between sister chromatid sets in telophase but not in anaphase, as if these cells had passed from anaphase to telophase prematurely, during the chromatid movement toward poles in anaphase A.     

Study of the testis in the active and hibernating hedgehog

The structure and ultrastructure of the hedgehog's testis has been studied during active and hibernating states. The maximum activity of the hedgehog's testis manifest itself in spring and summer. In autumn they enter in a seasonal resting, in which, the interstitial cells atrophy and it can observe primary spermatocytes resting next to spermatogonia and supporting cells in the seminiferous tubules. These have a minimal diameter and contain only primary spermatocytes in hibernating period also the seminiferous epithelium that compound the walls of the tubules are about 72 microns thick in the active period and about 47 microns in hibernation. The Sertoli cells shows a small diameter in hibernation than in the active season. The nucleus has one or two nucleoli with distinguish themselves from those observed in the active season because of their closer union with the inner nuclear envelope. In autumn (resting period) the interstitial tissue regress and the Leydig cells are aggregated in blocks with little cytoplasm, minimum size nuclei and 1 or 2 nucleolus closely to the inner nuclear membrane. Like then in the ovary, the hedgehog's testis follows a seasonal cycle determined by the temperature and environmental changes and it have a power of physiological adaptation inherent of a typical hibernating animal.     

Numerical analysis of the method of internal dialysis of giant axons.

This paper presents a numerical analysis of the method of internal dialysis used for studies of membrane transport in giant axons. Account is taken of the complete geometry, end effects, and finite dialyzate flow rates. Both influx and efflux experimental conditions are considered. Results place quantitative limits on system performance that are sufficiently general for use in experimental design. The completeness of solute equilibration and the uniformity of solute concentration at the axon membrane are assessed, as well as the sensitivity to dialysis solution flow rate. The effects of undialyzed axon ends on the equilibration rate and on the uniformity of concentration are determined, and the contribution of undialyzed ends to influx measurements is evaluated. Equilibration times are correlated with physical properties of axoplasm and dialysis tubing, and with dialysis solution flow rate. Numerical results confirm the general qualitative assessments of the method that are based on years of successful application.     

Water permeability of polyunsaturated lipid membranes measured by 17O NMR.

Diffusion-controlled water permeation across bilayers of polyunsaturated phospholipids was measured by 17O nuclear magnetic resonance. In 100-nm extruded liposomes containing 50 mM MnCl2, water exchange between internal and external solutions was monitored via changes in the linewidth of the 17O water resonance of external water. Liposome size and shape were characterized by light scattering methods and determination of liposome trapped volume. At 25 degrees C, the following water permeability coefficients were determined: 18:0-18:1n-9 PC, 155 +/- 24 microns/s; 18:0-18:3n-3 PC, 330 +/- 88 microns/s; and 18:0-22:6n-3 PC, 412 +/- 91 microns/s. The addition of 1 M ethanol reduced permeability coefficients to 66 +/- 15 microns/s for 18:0-18:1n-9 PC and to 239 +/- 67 microns/s for 18:0-22:6n-3 PC. Furthermore, the addition of 50 mol% 18:1n-9-18:1n-9 PE reduced the water permeability from 122 +/- 21 microns/s for pure 18:1n-9-18:1n-9 PC to 74 +/- 15 microns/s for the mixture. The significant increase in water permeation for membranes with polyunsaturated hydrocarbon chains correlates with looser packing of polyunsaturated lipids at the lipid-water interface and the suggested deeper penetration of water into these bilayers. Ethanol may block water diffusion pathways by occupying points of water entry into bilayers at the interface. The addition of dioleoylphosphatidylethanolamine increases lipid packing density and, consequently, reduces permeation rates.     

Visualizing the postembryonic development of Sarcophaga peregrina (flesh fly) by NMR microscopy

The postembryonic development of the flesh fly was studied using high resolution nuclear magnetic resonance imaging. Because this development occurs in a puparium, this process cannot be observed directly using standard histological techniques. The remodelling of histolysing larval tissues to developing imaginal tissues including the yellow body, a transient alimentary structure, and the integration of the developed adult structure were revealed in the images. Most surprisingly it was found that a large gas space that forms in the central region of the prepupa moves to the dorso-anterial region in less than 5 min due to the larval–pupal apolysis together with separation of the developing pupal epidermis from the puparium.