“The French guerrillas would always avoid confrontation. Instead, disguised as civilians, and without any distinguishing marks, they would lurk in the shadows, waiting for a chance to strike in some clandestine way . . .”
—Major Georg Grossjohann, 198th Division
The location of the Anvil landings was determined by two overriding factors: the need to secure ports on the French Riviera, specifically Toulon and Marseille, capable of sustaining a drive north up the Rhône Valley to link with the Allied forces pushing south from their lodgement in Normandy; and a fervent desire not to repeat the disaster at Anzio, where Wehrmacht control of the high ground beyond the beachhead kept the Allies under continual observation and enemy fire. This time the Anvil planners were determined to select a location where a landing force could quickly seize the high ground beyond the beachhead and establish a firm defensive line.
The planners’ first move was to study the coastline to the west and east of Toulon and Marseille as well as the coastline in between them. To the west, the area around Sète-Agde, south of Montpellier, was examined, as a landing there would permit a drive through the Carcassonne Gap and Toulouse toward Bordeaux. However, the planners concluded that poor beach gradients and the many lagoons, lakes, and canalized roads in the hinterland of the assault area would pose too many problems for the Allies. In addition this area was beyond the range of effective air cover from the nearest available Allied air bases in Corsica,* and intelligence had indicated that the approaches to the beaches were heavily laced with obstacles and other defenses. Clearly the Germans had selected this stretch of coast as the most probable point of attack.
The Golfe de Fos, a few miles northwest of Marseille, offered some promising landing beaches, but here the hinterland was flat and swampy and cut by the Rhône estuary. The bay at La Ciotat, a heavily fortified harbor equidistant from Marseille and Toulon, emerged as another candidate but would have exposed any assault force to heavy fire. The same drawback ruled out Hyères, east of Toulon, which could only be approached through shallow waters, restricting naval gunfire support, and was within the range of the port’s numerous heavy guns and also those that Allied intelligence believed were stationed on the offshore Hyères islands, the nearest of which was only seven miles away. Preliminary operations against these islands, Port-Cros and Le Levant, would alert the enemy to Allied intentions.
To the east of Hyères lay the Var coastline, not especially well defended but with beaches of widely varying size from 500 to 4,500 yards long and from 10 to 50 yards deep. Behind these beaches were narrow belts of sand dunes, many of which sported small scrub trees. Inland from the dunes were wooded slopes, cultivated fields, interspersed with areas of marsh and soft sand, small streams, farm tracks, and villages. A small macadam road ran along the coast here. Three miles inland was the principal east-west road from Toulon to Cannes, interrupted by a ten-mile section from Sainte-Maxime to Fréjus, where it joined the coastal road.
Rising behind the Toulon–Cannes road was the high ground of the Massif des Maures (Range of the Moors) that loomed above the coast from Hyères to Saint-Raphaël, a distance of some thirty-five miles, reaching a maximum height of 1,500 feet and stretching inland for twenty miles. Heavily wooded and rich in chestnut trees, maritime pines, cork oak, and dense maquis shrubbery, the massif offered the prospect of high ground, the rapid seizure of which was crucial to the Allied planners. To the north of the massif, was the main east-west corridor in the valley of the Argens River with good roads connecting the French Riviera with Marseille. Northeast of the corridor lay the Alpes de Provence that overlooked a smaller massif, the Esterel, which rises above Cannes.
The planners’ focus was now drawn to the coastline from Cavalaire east to Saint-Raphaël. Here seemed the most promising point of attack. The beaches on this stretch of coastline were deemed adequate, but no more, with satisfactory gradients and a deep water approach. The beach defenses were not as strong as those to the west, and the Argens River Valley offered the chance of a rapid move to the west to isolate the defenders of the cities of Marseille and Toulon. Nevertheless, the choice of the Cavalaire-Saint-Raphaël coast carried with it several significant drawbacks. Its thirty miles of irregular coastline translated into over fifty miles of shoreline. The potential landing sites were little more more than adequate and were cut off from each other by cliffs and rocky outcrops. Above them hung the Massif des Maures, posing a potentially fatal threat if it was held in strength by the enemy, a repeat of the Anzio debacle. Some of the prospective beaches lacked good exits, and these could be easily blocked by the German defenders, as could the narrow coastal road to Toulon and the highway from Saint-Raphaël to Toulon, which was only marginally wider. The minor roads that fed the main highway were, in large part, no more than a single lane wide and could accommodate only light military traffic.
The planners settled on the Cavalaire-Saint-Raphaël option, but the need to secure high ground as quickly as possible, and to rapidly develop supporting airfields, obliged the Seventh Army’s planners to map out a large beachhead with a radius of some twenty miles, at the center of which was Cap Saint-Tropez. Twenty miles to the west, the perimeter of the beachhead, dubbed the “Blue Line,” rose from the Rade d’Hyères, snaking through the western tip of the Massif des Maures before looping east just north of Le Luc, near the middle of the Toulon-Saint-Raphaël corridor, and running on past Taradeau and Le Muy, the latter some ten miles inland on the Argens River, from which a good cement road ran down to Saint-Raphaël. Then it curved gently eastward through the smaller Massif de l’Esterel down to the sea at the Golfe de la Napoule, two miles northeast of the town of Théoule. Inside the Blue Line lay important sections of the principal highways and railways leading west and southwest to Toulon, Marseille, the Rhône River, as well as northeast to the cities of Cannes and Nice. The planners were confident that the depth of the beachhead would protect the assault beaches from long-range German artillery; provide space for the development of airfields; and allow U.S. 6th Corps and the leading elements of French 2nd Corps the freedom to maneuver. The two massifs, once taken, would secure the invasion force’s western and eastern flanks against a determined enemy counterattack while preparations were made to break out west toward Toulon and Marseille.
The landings by 6th Corps were to be preceded by four key missions: the seizure of the offshore islands of Port-Cros and Le Levant by French commandos and the U.S. First Special Service Force (Operation “Sitka”) to neutralize the German batteries thought to be there; assaults by the French African Commando Group (Operation “Romeo”) on the mainland north of the islands at Cap Nègre to destroy coastal defenses, establish roadblocks on the coastal highway, secure the high ground up to two miles inland, and block any German attempt to move on the landing beaches; air drops by parachute troops of First Airborne Task Force (Operation “Rugby”) around Le Muy aimed at preventing German attempts to mount attacks on the landing beaches from the Argens River corridor. The paratroops were also tasked with securing the area for the glider landings that would follow later that day before joining hands with advanced elements of 6th Corps coming up from the coast. And, finally, on the eastern end of the beach-line, the French Naval Assault Group (Rosie Force) was to block the coast highway on the right, or northeastern flank of 6th Corps. H-Hour for the landings by 6th Corps was set for 0800, markedly later than previous amphibious operations in the Mediterranean. This was to give the pre-assault air and naval forces sufficient daylight in which to reduce the enemy’s shore defenses, following which the assault forces could establish themselves ashore and seize the high ground behind the landing beaches to forestall a German counterattack. Particular importance was attached to the formation of twenty-four naval shore fire control parties (NSFCPs) to coordinate fire from the Task Force warships. There was to be a fire control party for each of the eighteen assault battalions, consisting of officers with combat experience gained during the Normandy landings.
In the original plan, 45th Division was positioned in the center, attacking beaches to the northeast of Saint-Tropez; 36th Division was on the left, landing on beaches to the southwest of Saint-Tropez; and 3rd Division was on the right, targeting beaches to the northeast of Fréjus. When these plans were submitted to Truscott toward the end of June, he raised several pertinent objections. He insisted that his most experienced division, the 3rd, should be shifted to the left (Alpha) flank, on the beaches designated 259, 260, 260-A, 261, and 261-A, to launch a swift drive on Toulon or parry a determined enemy counterattack from that quarter. The 36th, which was less experienced, should be on the right (Camel) flank, landing on beaches 264, 264-A, 264-B, 265, and 265-A, to the northeast of Fréjus and deployed in a largely defensive role. The 45th retained its position in the center of the assault landing, concentrating on the (Delta) beaches to the northeast of Sainte-Maxime, 262, 262-A, and 263-A, B, and C. Each division was to have an attached tank battalion, a tank destroyer battalion, three antiaircraft and artillery battalions, a barrage balloon battalion plus three battalions of corps artillery. In addition there was provision for engineers, hospitals, and supply and maintenance troops.
Truscott was also worried that the westward extension of the beachhead ran the risk of repeating the mistake of Anzio, preventing the rapid massing of a force for the westward drive to Toulon. He suggested that the Blue Line to the west be pulled in to accommodate this maneuver. Patch and Admiral Hewitt agreed to these modifications, but Truscott failed to persuade Hewitt to mount a comprehensive reconnaissance sweep for underwater obstacles sown by the enemy in the approaches to the assault beaches. Hewitt was convinced that, in contrast to the Normandy landing beaches, the low tidal range on the coastline chosen for Anvil posed little danger. Moreover, he reminded Truscott that reconnaissance operations carried the risk of alerting the enemy to the landing sites selected by the Americans. Truscott continued to fret about underwater obstructions, and his fears were raised by intelligence reports of such obstructions on the approaches to at least three of the assault beaches. He raised the matter again with Hewitt, who replied, “I believe that the menace of underwater obstacles has been somewhat exaggerated,” and that “studied reports from Overlord have lent undue emphasis to a subject new in this theater for the most part . . . I know of no preliminary reconnaissance other than actually running boats through the obstacles, which will ensure that boats can beach.” Hewitt rejected the suggestion that pontoon causeways could be used to smash through these defenses but agreed that LCTs (Landing Craft, Tank) could undertake the task, coupled with hand demolitions after the first wave had landed.
Truscott was nothing if not persistent. He dispatched another memorandum: “Without exception every Navy officer with whom I have discussed the subject has expressed the opinion that underwater obstacles constitute a more serious problem for us than was the case in the [English] Channel where the tidal range allowed assault craft to beach and allowed demolition parties to work on dry land. The available information on the nature of the obstacles in our area does not indicate that these obstacles are exceedingly formidable, however, I think that we must assume that these obstacles do have the capability of interfering with the approach of the landing craft to the beaches to some degree at least, and certainly the accomplishment of the troops’ mission on the shore depends upon their reaching the beach.”1 Truscott also pointed out that two of Hewitt’s subordinate Sub Task Force commanders, Admirals Lowry and Rodgers, had proposed to make an early reconnaissance by speedboat several hours before H-Hour if Hewitt gave the go-ahead. The approach worked. Hewitt agreed to leave the solution of the problem to his subordinates, who would also consult with Truscott’s divisional commanders.
Truscott was also preoccupied with 6th Corps’s lack of an American armored combat command with which to expand and burst out of the bridgehead. The concept of an armored combat command had been introduced into the U.S. Army in March 1942. It was essentially a headquarters, with no organic combat units, which controlled the armored elements assigned to it for a specific tactical operation, for example the support of an infantry division. The French Army, rebuilt and equipped by the Americans, also featured combat commands (CCs), and one of these, the French 1st Armored Division’s Combat Command Sudre, commanded by Brigadier General Aimé M. Sudre, was attached to 6th Corps. CC Sudre comprised a reconnaissance squadron, a regiment of heavy tanks, a battalion of motorized infantry, a group of motorized artillery, a squadron of tank destroyers, engineers, signals units, and elements of transport and service corps—a total of over 3,000 men and some 1,000 vehicles. Each French armored division was divided into three combat commands, any one of which could fight independently from its parent formation in close cooperation with infantry formations.
Truscott hoped to land CC Sudre over 36th Division’s beaches around Fréjus and concentrate it for a drive to the northwest. This radically altered the approach to Dragoon, replacing the slow and methodical buildup and progress inland envisaged by the planners with a more ambitious, but potentially riskier, rapid exploitation following the landings. However, the French expected to regain control of CC Sudre on D+3 to shield their right flank from attacks by German armored units as they moved on Toulon and Marseille. This forced Truscott to improvise a solution. On August 1, the day the code name for Anvil was changed to Dragoon, he told his planning staff that he had decided to assemble a provisional armored reserve from 6th Corps assets. It was to be led by Brigadier General Fred B. Butler, 6th Corps’s deputy commander, and was designated Task Force Butler (TFB). Because Task Force Butler was almost an afterthought, it could not be added to the invasion force as a separate entity and was scheduled to be assembled in the landing area on D+2.
Butler was a graduate of West Point who had trained as an engineer. He had seen combat commanding 168th Regiment, 34th Infantry Division, in North Africa and Italy. In 1944 he was promoted to the rank of brigadier general and was reassigned to 34th Division as its assistant commander. As Truscott’s deputy in 6th Corps, he had been at Anzio, where he became familiar with all his senior’s views and, according to Truscott, was “one of the most fearless men I ever met.” Butler selected many of his staff from 6th Corps’s staff sections. They were veterans of Anzio and the subsequent breakout.
Task Force Butler was to be built around 6th Corps’s mechanized cavalry formation, the 117th Cavalry Reconnaissance Squadron,† whose 900 troops rode in jeeps mounting .50 machine guns and 60mm mortars, and forty M8 Greyhound armored cars. Developed by the Ford Motor Company, the 6x6 rear-engined, seven-and-a-half-ton Greyhound entered service with the U.S. Army in 1943. It was quiet and fast, with a top speed of 55 mph, but was vulnerable to mine damage, making it unpopular with its four-man crews, who reinforced its floor with sandbags. Its main armament was a 37mm gun and it also carried a .30 machine gun and a turret-mounted .50 machine gun. The Greyhound did not have the firepower to destroy German heavy or medium tanks, but was effective against less heavily armored vehicles such as half-tracks or armored cars, soft-skinned vehicles, and personnel.
Owing to its dedicated role of reconnaissance, the Greyhound enjoyed a superior communications system. Most Greyhounds carried two radios, one short-range and one long-range, which would in theory enable Butler to communicate with higher headquarters and subordinates, either directly or by relaying transmissions over long distances. Nevertheless, the mountainous interior of southern France would pose a number of problems for rapid and regular radio communication. Butler’s Greyhounds may have lacked armored punch but conferred on TFB the ability to locate and report on enemy strength, bypass enemy concentrations, and expand the operational depth and frontage on which it could function in a fast-moving battle.
The 117th Reconnaissance Squadron also fielded a single assault gun troop equipped with six M8 75mm Howitzer Motor Carriages. These assault guns were tracked vehicles that could be used in a number of roles. They were particularly effective in attacking enemy positions in built-up areas and could also be used to provide indirect fire support. The reconnaissance squadron also had an organic company of seventeen M3 and M5A1 light tanks. However, it was in the provision of medium tanks that TFB’s profile did not match that of the standard combat command. It had only two companies of medium tanks (from the 753rd Tank Battalion) each with seventeen M4 Shermans armed with a 75mm gun, plus one M4 mounting a 105mm howitzer that gave close support to medium tank formations. To offset this deficiency in firepower, TFB was assigned a company of twelve open-topped M10 tank destroyers (Company C, 636th Tank Destroyer [TD] Battalion) armed with a 3-inch high-velocity gun in a turret capable of an allround traverse. The M10 was not a true tank but was often used in this role when no M4 Shermans were available. More indirect fire support was provided by the self-propelled field artillery battalion (59th Armored Field Artillery) attached to Task Force Butler. The battalion was equipped with eighteen M7 open-topped Howitzer Motor Carriages, carrying a 105mm howitzer and a pulpit-mounted Browning .50 antiaircraft machine gun to the right of the main armament.
Task Force Butler was also assigned a single motorized infantry battalion (2nd Battalion, 143rd Regiment) riding in two-and-a-half ton (deuce-and-a-half) trucks rather than the more maneuverable M3 half-tracks used by units organic to armored divisions. Half-tracks could follow tanks wherever they went, but Butler’s trucks were relegated to the roads. The motorized infantry deployed fewer machine guns than their armored counterparts but were equipped with more mortars, the most responsive indirect fire available to commanders in the field. Moreover, in heavily forested or mountainous terrain, as was the case in the South of France, tanks were often forced to advance by road, losing their cross-country mobility. When one compares the inventories of an American combat command and that assembled at extremely short notice by Brigadier General Butler, one can only conclude that his organizational skills had created a formation with capabilities that exceeded those of a regular armored combat command, not least in the conduct of reconnaissance and security missions. In all, Task Force Butler fielded some 3,000 men and 1,000 fighting vehicles.
At the beginning of July, Truscott reflected that “Our plans were based upon a thorough knowledge of terrain and beaches in southern France, particularly of the assault area, and on an unusually complete intelligence concerning enemy strength and dispositions. Not even the Normandy invasion had better advance information.”2 This information principally flowed from three distinct sources: aerial reconnaissance, the French Resistance, and Ultra, the last the code name for the decrypts of intercepted German radio traffic, made at Bletchley Park and forwarded to senior Allied commanders and their staffs in the field. Ultra-derived information gave Allied intelligence an overall picture of German capabilities and dispositions in the run-up to Dragoon, but the speed of German redeployments after Overlord, combined with the time-lag inevitably involved in the process of decryption and its correlation with other intelligence—prisoner of war interrogations, air mosaic analysis, captured documents, Resistance reports—sometimes clouded the Allied assessment of precise enemy intentions. Moreover, the German practice of making many key decisions at military conferences, and the transmission of detailed orders by officers entrusted with documents, made Ultra a less than perfect, albeit indispensable, tool in the Allies’ intelligence toolbox. It was rare that an Ultra decrypt had an immediate effect on an unfolding battle, as happened at Mortain in early August 1944. Nevertheless, a series of intercepts after the August 15 landings were to prove critical to the course of the campaign.
Old-fashioned human intelligence (HUMINT), derived from agents on the ground, provided an alternative to Ultra’s insights. Here the OSS came into its own. One of their outstanding agents, based in Algiers, was Henry Hyde, born in Paris and the scion of a wealthy American family, who had been educated at the British public school Charterhouse and Cambridge University. Hyde was the epitome of the sarcastic OSS sobriquet “Oh, So Social,” and had been recruited in 1942 by spymaster Allen Dulles, who later became the director of the Central Intelligence Agency (CIA). From Algiers, Hyde ran the “Penny Farthing” network, which from July 1943 controlled agents, ranging from anarchists to aristocrats, in northern and southern France. In February 1944, Hyde spotted that the German 2nd Panzer Division had arrived in southwest France to refit, but had not been picked up by Ultra. In early June 1944, informed by Hyde, members of the Resistance had disabled much of the division’s replenished armor and delayed its arrival in Normandy by two weeks. By the late spring of 1944, Hyde had established an extensive network of agents in the South of France, supplying a mass of information by radio and maps, diagrams, drawings, and photographs smuggled through Spain. His agents also accumulated significant amounts of information about roads, bridges, and German installations and troop movements, all of which were supplied to the Anvil planners. A final piece of the intelligence jigsaw came from a wealth of prewar tourist photographs of the assault beaches, which provided planners with detailed insights into the target area as significant landmarks loomed behind the tanned, smiling faces of vacationers.
Up to and during the landings, the Seventh Army’s G-2 (Intelligence), Colonel William W. Quinn, ensured that he received a constant stream of information from OSS reports on enemy dispositions and strength. On August 13, two days before the landings, a French OSS agent, bicycling from Cannes to Hyères, was able to make a detailed survey of the landing areas, which was cabled to Quinn aboard the Seventh Army’s command ship, Catoctin. After the landings, the OSS assigned teams to each of the three US divisions involved in Dragoon to provide information on the enemy’s order of battle.
In the weeks preceding the Allied landings, the FFI in the Var was encouraged to do all it could to hamper German movement into the Anvil assault area by destroying bridges, cutting railway lines, blocking roads, and creating major diversions inland after August 15 to distract enemy forces maneuvering to counterattack. On the face of it, the prospects looked promising. The FTP was active throughout the region; the ORA was particularly strong in the northwest of the department; the Gaullist AS operated principally in Toulon and Draguignan, the department’s administrative center; and by mid-August many of these groups had joined forces to form the CFL (Corps Français de La Libération).
The most successful unification was achieved in the Massif des Maures, where the Brigade des Maures was created by a young architect, Marc Rainaud, who had forged an alliance between members of the FTP, the ORA, and the AS, merging them with local Maquisard and Resistance groups farther afield in the Saint-Tropez peninsula. However, Rainaud was not informed that the 3rd Division was going to land on beaches to the west of the peninsula. In addition, Allied attempts to supply the Brigade des Maures with air-dropped containers and packages were sporadic. After June 6, only eight deliveries were successful, and one of these occurred after August 15. In the days immediately leading up to the Dragoon landings, frantic efforts were made to step up the landing of agents and teams who had been training in Algiers into zones just beyond the Allied Blue Line. One of the agents was Lieutenant Geoffrey M. T. Jones, a twenty-four-year-old, French-speaking OSS training officer who had been instructing recruits at Blida airfield, outside Algiers, since June. He was accompanied by a French naval officer Capitaine de Corvette Léon Allain, the leader of a counter-scorch team bound for Toulon with orders to open the port for Allied shipping.
At this stage in the war the Germans seldom ventured in force into the Provençal countryside, but they could still make life difficult for the Resistance active in major conurbations. In the spring of 1944 an entire network had been penetrated in Nice, in the Alpes Maritimes, the department adjoining the Var. German countermeasures were particularly effective in Marseille, where a number of Resistance leaders had been identified by colleagues “turned” by SIPO/SD operative Ernst Dunker, a shady trilingual character with a prewar background in international catering. The Résistants were arrested in two raids, at Oraison in the Basses-Alpes, and Marseille itself. On July 18, twenty-six Résistants were taken to a lonely spot near the hamlet of Signes, some fifteen miles north of Toulon. There, in an unfrequented copse, they were shot and buried in a shallow grave. Just over a month later, their corpses were joined by ten more executed Résistants.
In the remote countryside of southern France, the Resistance was less vulnerable. In the hilly, forested regions of the southeast, the Maquis had so rattled the Germans that, with the exception of anti-guerrilla operations, like that in the Vercors, and heavily escorted convoys, they rarely ventured beyond the relative safety of their urban garrisons, leaving tracts of countryside open to bands of Résistants. The corrosive mixture of fear and contempt felt by many German soldiers toward the Resistance was captured by Georg Grossjohann in his memoirs: “In my experience, these guerrillas only appeared where and when they could do so. Typically they would liquidate small German rearguards or scattered groups, and then mostly from ambush. Even then, the German OKW allowed them the status of combatants if they complied with two conditions—to come forward as an integral military unit and to wear clearly visible armbands to identify themselves. Nevertheless, the French guerrillas would always avoid open confrontation. Instead, disguised as civilians, and without any distinguishing mark, they would lurk in the shadows, waiting for a chance to strike in some clandestine way . . . German soldiers disappeared again and again without a trace. The most common victims of French ‘resistance’ included lone messengers or communications people who were sent out in small parties to repair cables.”3 Later in his memoirs Grossjohann records his grim satisfaction at the summary execution of an armed Résistant reconnoitering his position without displaying a white armband marked with the Cross of Lorraine insignia of the FFI. Grossjohann signally failed to comprehend that when the Resistance went toe-to-toe with well-armed and disciplined enemy troops, it always came off second-best. Nevertheless, the drip-drip effect of the hit-and-run tactics he found so contemptible steadily undermined German morale.
As Army Group G’s strength diminished with the transfer northward of many of its most capable units, the FFI became increasingly emboldened, threatening the Germans’ two most important lines of communication, the Carcassonne Gap and the Rhône Valley. The operation to secure the upper Rhône Valley, the harrowing of the Vercors, proved only a partial success as it prompted an explosion of brush fires in other parts of southern France. Sabotage of roads, railways, and bridges reached epidemic proportions. In the first two weeks of August, the FFI cut the railway lines in the Carcassonne Gap and the Rhône Valley some forty times and damaged or destroyed thirty-two road and rail bridges. The cutting of underground and overhead telephone and telegraph lines shredded Army Group G’s communications with its forces on the Atlantic coast and OB West. Because the mountains in southern France significantly interfered with radios—a phenomenon also endured by the Americans after the landings of August 15—Army Group G resorted to going through Berlin rather than OB West in Paris. However, the increased radio traffic was now rendered more vulnerable to interception and subsequent decryption at Bletchley Park. A week before the landings, Blaskowitz complained that the Resistance was no longer a negligible terrorist threat but constituted an organized and full-blown army threatening his rear.
The degrading of German road and rail communications was a task also undertaken principally by the Allied air forces in the weeks preceding the Dragoon landings. By August 15, MAAF had destroyed almost all the important rail and highway bridges over the Rhône, Durance, and Var Rivers, severely disrupting communications with Lyon. This came at a heavy cost of French civilian lives. One example will suffice. On August 13 the Americans had bombed the railway bridge at Crest, missing the bridge—which in any case had been rendered useless by the Resistance two months earlier—but killing 530 civilians, wounding another 713, and destroying 1,221 buildings.
In July, the weight of the Allied air offensive had been thrown against Italy and the Balkans, but thereafter the priorities switched to the South of France. The bridges on the Var were particularly badly hit. Josef Kirsner, with 164th Battalion, 148th Reserve Division, was a member of the detail guarding the western side of the Var bridge in Saint-Laurent. “There were some Italian soldiers or workers who were constantly repairing the bridge after it had been damaged in air raids so that cars and trucks could keep on using it. They had a trumpet, and whenever an air raid was coming they blew the trumpet before our own Fliegeralarm [air raid siren] sounded. Beside our house in the orchard we had dug real foxholes, almost two meters deep. When there was an air raid we ran like crazy and jumped into these holes. It was bit safer there than on the bridge. Once there was a train stopped in the train station at Saint-Laurent and they bombed it. It was only a couple of hundred meters away from where our foxholes were.”4
The bane of Kirsner’s life was an NCO from the battalion headquarters who was constantly checking the bridge guard to see if they were sleeping. The NCO insisted that Kirsner and his comrades take shelter during a raid in a flimsy Italian bunker built on the bridge. The detail members, preferring the relative safety of the shelter in the orchard, refused to use the bridge shelter. The NCO, setting an example, made a point of using the shelter, which, soon enough, took a direct hit and was blown into the river below. Kirsner recalled, “We practically found nothing of him. He died in his bunker and that was the end of him. Other than that we had no losses.”
Between April 28 and August 10, MAAF had dropped over 19,500 tons of bombs on southern France. The second phase of the air effort, Operation “Nutmeg,” began on August 10 and targeted enemy coastal batteries and defense formations with the overall aim of isolating the invasion area. To retain tactical surprise, an almost impossible task, the aerial bombardment stretched over the Mediterranean coastline in an arc from the Spanish border to Viareggio in Italy.
On the ground in the Dragoon target area, one particularly important operation was launched by the OSS man Geoffrey Jones and his French colleague Capitaine Allain, both of whom had made a series of successful rendezvous with Maquis leaders and some two hundred Maquisards at Draguignan, Montauroux, and Les Arcs. On the afternoon of August 14, less than twenty-four hours before the launching of Dragoon, Jones, Allain, and the Maquis leaders met at the village of Mons, thirty miles from Draguignan, where at 2000 they received a BBC message, “Nancy a le torticule” (Nancy has a stiff neck) indicating that Dragoon was imminent.
Jones and Allain decided to knock out the German radar installation at Fayence, a “village perché,” overlooking the plain south of the Esterel Massif. The radar had been erected at an abandoned reservoir above Fayence with its antennae perched atop a huge boulder from which it could survey the Allied landing beaches at Saint-Raphaël, some fifteen miles away. Jones, who had heard about the radar in Algeria, was well aware of its importance, and with his Resistance team crawled up the old reservoir’s water mains, placed explosives, and blew it up. Jones’s and Allain’s next task was to proceed to the landing zone of First Airborne Task Force and to prepare it for the arrival of Frederick’s airborne American and British troops.