* Laboratoire de physiologie, Faculté de Médecine Saint-Antoine, Paris, France
As observed at the turn of the century by Marks & Spencer (1899), who first named the ``yelling reaction'' (YR), the striking effects of tomato throwing on Sopranoes have been extensively described. Although numerous behavioral (Zeeg & Puss, 1931; Roux & Combaluzier, 1932; Sinon et al., 1948), pathological (Hun & Deu, 1960), comparative (Karybb & Szyla, 1973) and follow-up (Else & Vire, 1974) studies have permitted a valuable description of these typical responses, neuroanatomical, as well as neurophysiological data, are, in spite of their number, surprisingly confusing. In their henceforth late twenties' classical demonstrations, Chou & Lai (1927 a, b, c, 1928 a, b, 1929 a, 1930) have ruled out the hypothesis of a pure facio-facial nociceptive reflex that has been advanced for many years by a number of authors (Mace & Doyne, 1912; Payre & Tairnelle, 1916; Sornette & Billevayzé, 1925). Since that time, numerous observations have been made that have tried to decipher the tangling puzzle as well as the puzzling tangle of the afferent and/or efferent sides of the YR and led to the rather chaotic involvement of numberless structures and paths: trigeminal (Loewenstein et al., 1930), bitrigeminal (Von Aitick, 1940), quadritrigeminal (Van der Deder, 1950), supra-, infra-, and inter-trigeminal (Mason & Ragoun, 1960) afferents have been likely pointed out as well as macular (Zakouski, 1954), saccular (Bortsch, 1955), utricular (Malosol, 1956), ventricular (Tarama, 1957), monocular (Zubrowska, 1958), binocular (Chachlik, 1959-1960), triocular (Strogonoff, 1960), auditive (Balalaika, 1515) and digestive (Alka-Seltzer, 1815) inputs. Spinothalamic (Attou & Ratathou, 1974), rubrospinal (Maotz & Toung, 1973), nigro-suiatal (Szentagothai, 1972), reticular (Pompeiano et al., 1971), hypothalamic (Hubel & Wiesel, 1970), mesolimbic (Kuffler, 1969) and cerebellar (High & Low, 1968) pathways have been vainly searched out for a tentative explanation of the YR organization and almost every part of the somesthesic (Pericoloso & Sporgersi, 1973), motor (Ford, 1930), commissural (Gordon & Bogen, 1974) and associative (Einstein et al., 1974) cortices have been found responsible for the progressive building-up of the response although, up-to-now, no decisive demonstration of both the input and output of the YR programming has been convincely advanced.
Recent observations by Unsofort & Tchetera pointing out that ``the more you throw tomatoes on Sopranoes, the more they yell'' and comparative studies dealing with the gasp-reaction (Otis & Pifre, 1964), hiccup (Carpentier & Fialip, 1964), cat purring (Remmers & Gautier, 1972), HM reflex (Vincent et al., 1976), ventriloquy (McCulloch et al., 1964), shriek, scream, shrill and other hysterical reactions (Sturm & Drang, 1973) provoked by tomato as well as cabbages, apples, cream tarts, shoes, buts and anvil throwing (Harvar & Mercy, 1973) have led to the steady assumption of a positive feedback organization of the YR based upon a semilinear quadristable multi-switching interdigitation of neuronal sub-networks functioning en desordre (Beulott et al., 1974). Although this hypothesis seems rather seductive, it lacks anatomical and physiological foundations and we therefore decide to explore systematically the internal incremental or decremental organization of the YR, allowing a tentative anatomic model.
Experiments were carried out on 107 female healthy Sopranoes (Cantatrix sopranica L.) furnished by the Conservatoire national de Musique, and weighing 94-124 kg (mean weight: 101 kg). Halothane anesthesia was utilized during the course of tracheotomy, fixation in the Horsley-Clarke, and major operative procedures. 5 % procaine was injected into skin margins and pressure points. Animals were then immobilized with gallamine triethyiodide (40 mg/kg/hr) and normocapnia was maintained by appropriate artificial ventilation. Spinal cord transections were performed at L3/T2 levels, thus eliminating blood pressure variations and adrenaline secretion induced by tomato throwing (Giscard d'Estaing, 1974). The fact that the animals were not suffering from pain was shown by their constant smiling throughout the experiments. Internal temperature was maintained at 38 °C ± 4 °F by means of three electrically drived boiling kettles.
Tomatoes (Tomato rungisia vulgaris) were thrown by an automatic tomatothrower (Wait & See, 1972) monitored by an all-purpose laboratory computer (DID/92/85/P/331) operated on-line. Repetitive throwing allowed up to 9 projections per sec, thus mimicking the physiological conditions encountered by Sopranoes and other Singers on stage (Tebaldi, 1953). Care was taken to avoid missed projections on upper and/or lower limbs, trunk & buttocks. Only tomatoes affecting faces and necks were taken into account. Control experiments were made with other projectiles, as apple cores, cabbage runts, hats, roses, pumpkins, bullets, and ketchup (Heinz, 1952).
Unit activity was recorded through glasstungsten semi-macroelectrodes located au-petit-bonheur, according to the methods of Zyszytrakyczywsz-Sekrâwszkiwcz (1974). Spike recognition was performed by audiomonitoring: every time a unit discharge was heard, it was carefully photographed, tapped, displayed on a monograph and, after integration, on a polygraph. Statistical evaluation of the results was made using a tennis like algorithm (Wimbledon, 1974), that is, every time a structure responds up to win the game, it was recognized as YR-related.
At the end of the experiments, Sopranoes were perfused with olive oil, and 10 % GlennFiddish, and incubated at 421 °C in 15 % orange juice during 47 hours. Frozen 2 cm unstained sections were mounted into delta-strawberry sherbet and observed under light and heavy microscopy. Histological verifications confirmed that all the electrodes were located in the brain except four that were found in cauda equina and filum terminale and disclosed from statistical analysis.
Stereotaxic explorations of brains during tomato throwing showed that most of the areas respond differently to the tomesthetic stimulation. As can be seen from table one, where the results are summarized, three (3) distinct areas gave definite, unambiguous and constant responses: the nucleus anterior reticularis thalami pars lateralis (NARTpl), or nucleus of Pesch (Pesch, 1876; Poissy, 1880; Jeanpace & Desmeyeurs, 1932), the anterior portion of the tractus leguminosus (apTL), lying 3.5 mm above the obex and 4 mm right of the tentorium and the dorsal part of the so-called ``musical sulcus'' (scMS) of the left hemisphere (Donen & Kelly, 1956). It is of interest to notice that, if the left hemisphere was kept for analysis, the right hemisphere was left.
Examples of responses obtained from these structures can be seen on figure 1 where temporal analysis of the spike distribution based on their Responsive-Area-Temporal-Programming (RATP) properties allowed to distinguish 3 unit subtypes: 1) units responding before the stimulation; 2) units responding during the stimulation and 3) units responding after the stimulation.
| Regions | Tomatic stimulation | |||||||||
| 1/s | 2/s | 3/s | 4/s | 5/s | 15/s | |||||
| whole brain | 0.0 | 0.0 | 4.2 | 0.6 | 0.7 | 000.1 | ||||
| raphe area | 3.1 | 4.1 | 5.9 | 5.9 | 5.9 | 000.2 | ||||
| septum | ±1 | 67 | 875 | 121 | 000 | p3517 | ||||
| thalamus | 2.2 | Ö3 | 456 | ±7 | 8.9 | 0.0001 | ||||
| NARTpl | 456 | +2 | -4 | $$ | " 2 " | ±0.001 | ||||
| hypothalamus | ±¢¢3 | 1&2 | 41 | S.G. | 121 | many | ||||
| hippocampus | 1/2 | 3% | Ö[S7] | ? | < 16 | 0±±7 | ||||
| cereb. cortex | yes | < 55 | nsp | {
| ±¥ | 71±70 | ||||
| scMS | ~ 31 | ~ 35 | > 87 | 00+ | [345/{ 4} ] | a few | ||||
| apTL | 0.0 | 3.1 | 6.7 | Ö4 | - | 56% | ||||
| amygdala | +3 | 0.0 | 4.2 | 0.6 | 0.7 | 000.1 | ||||
| N. Poissy | ® 8 | 0.0 | ® 1 | 12¬ | M/5 | 1+1=2 | ||||
| N. Pesch | 3$4 | 781 | 2 | ¯ 34 | ! | !!!!! | ||||
| N. ruber | D51 | ??? | å43 | ò07 | 415 | maybe | ||||
Cross-examination of responses driven by other projectiles and Ketchup stimulation are shown on figure 2 and argue unquestionably in favor of a tomatotopic organization of the YR along, between and across the NARTpl, apTL and scMS. Temporal relationships of those responses, as examplified in fig. 3, showed that the hypothesis of a clustering interdigitation of neuronal subnets is highly probable, although no experimental evidence can be given due to the relative difficulty of entering those damned structures without destroying a lot of things (Timeo et al., 1971).
It has been shown above that tomato throwing provokes, along with a few other motor, visual, vegetative and behavioral reactions, neuronal responses in 3 distinctive brain areas: the nucleus anterior reticular thalami, pars lateralis (NARTpl), the anterior portion of the tractus leguminosus (apTL) and the dorsal part of the so-called musical sulcus (scMS). As pointed out by Chou & Lai (1929 b), Lai & Chou (1931 a, b) and Unsofort & Tchetera (1972), the YR organization cannot be simply reduced to an oligosynaptic facio-facial nociceptive reflex which would have relayed over in the fascia leguminosa of the VIth laminations of the ventral quadrants of the paleospino-rubro-yello-tectocerebellonigrostriatal tomatonergic ascending pathways. For the fact that horseradish peroxidase injected into the Sopranoes' vocal cords is retrogradually transported from the apical dendrites of the vagus nerves to the tomatotomatic synapses of thc contralateral pseudogasserian afferents (McHulott et al., 1975) proves with some likelihood the leguminous nature of the mediator responsible for the transmission of the message from the receptive tomato fields to the YR circuitry (Colle et al., 1973). Thus, 3,5 (M-tri) argyril-beta-L-tomatase which is selectively trisynthetized in the NARTpl-apTL bundle and whose destruction blocks up drastically the YR (Others et al., 1974) stands out as the major candidate for the transmitter involved in the YR retroacting loop, although an alternate hypothesis based upon latency calculations, and cocross frequency correlations, puts forward the feasibility of a tomatotonic synapse (see Dendritt & Haxon, 1975). Although decisive experimental evidences are still lacking and further series of experiment are needed before the complete elucidation of the YR can be achieved, it seems logical to advance that above combined arguments along with experimental results described in our work are likely to support the hypothesis of a semi-linear multi-stable multi-switching net-back feedwork organization of the YR whose a tentative anatomical model can therefore be proposed (fig. 4).
This work was supported by grants from the Syndicat régional des Producteurs de Fruits & Légumes, the Association française des Amateurs d'Art Lyrique (AFAAL) and the Fédération internationale des Dactylo-Bibliographes (FIDB). The author gratefully acknowledges the helpful criticisms as well as the skilful assistance of J. Chandelier, M. De Miroschedji and H. Gautier.
Black lines = inhibitory; broken lines = interrogatory; dashed lines = redhibitory; stellate lines = whig-and-tory.