Center
of Military History
The
Corps of Engineers
The War Against
Germany and Italy
CHAPTER
X
The
Advance to the Alps
(Go
directly to 313th
info)
By
the time the Allied
armies collided
with the German
Winter Line defenses
in late 1943, the
American theater
command had changed
considerably. In
the aftermath of
the North African
invasion the need
to reorganize had
been clear; the
issue of new command
arrangements was
a lively one at
the American headquarters,
but the demands
of combat kept it
pending until the
downfall of Axis
forces in Tunisia
and Sicily.
The
chief defect still
lay in the over
lapping and sometimes
contradictory authorities
in the administrative
and supply chain.
A new theater engineer,
Brig. Gen. Dabney
O. Elliott, continued
to exercise his
advisory and staff
functions in three
separate commands—AFHQ;
NATOUSA; and COMZ,
NATOUSA—an
arrangement that
bypassed the Services
of Supply command.
No formal controls
of the engineering
function existed
between SOS, NATOUSA,
and the chief engineer
of the theater as
they did in General
Lee's SOS, ETOUSA,
jurisdiction in
the United Kingdom.
Maj.
Gen. Thomas B. Larkin
as chief of the
SOS, NATOUSA, command
had only nominal
control over the
base sections then
existing in the
theater and virtually
no say in the flow
of supply once materiel
moved out of the
bases for the front
lines. Larkin's
relationship with
the AFHQ G-4 was
unclear and in many
ways duplicative
through the period
of operations in
North Africa; it
improved only after
his concerted efforts
to revise the command
situation met with
some success.
Reorganization
In
March 1943, one
month after the
formation of the
theater, General
Larkin began a campaign
to eliminate the
anomalies and duplications
that weakened or
destroyed his effectiveness
as supposed chief
of all American
supply operations
in the theater.
He made small headway
against the resistance
of the staff officers
at NATOUSA and AFHQ
who insisted upon
retaining their
acquired authority,
citing in their
own behalf the dangers
of repeating the
bitter disputes
over the SOS, ETOUSA,
empire under General
Lee.
In
hopes of reducing
the manpower drains
in theater level
headquarters, the
War Department sent
an Inspector General's
survey team to North
Africa and to England
in late spring 1943.
The team's report,
in effect, recommended
a 50 percent reduction
in the number of
overhead personnel
in the theater staffs
in NATOUSA, a solid
impetus for reorganization
and economy in manpower.
Various
plans originating
at AFHQ and NATOUSA
undertook to eliminate
the command discrepancies
and to reduce the
manpower surpluses
in head quarters'
staffs. Their authors
usually proceeded
on the assumption
that vast changes
were necessary in
any staff element
but their own.
After
a summer and fall
of conflicting suggestions
in 1943, the SOS,
NATOUSA, command
had no increased
authority to deal
with its increased
responsibilities,
which now spanned
the Mediterranean
and extended to
a new base section
in Italy. Headquarters,
NATOUSA, insisted
upon the continued
control of personnel
in the base sections,
denying to Larkin
efficient use of
manpower and timely
use of specialty
units when he needed
them.
The
arrival of a new
theater commander
broke the impasse
and presaged
the decline of Headquarters,
NATOUSA, and the
disappearance of
COMZ, NATOUSA, in
early 1944. On 31
December 1943, Lt.
Gen. Jacob L. Devers
relieved General
Eisenhower, who
returned to ETOUSA.
When Devers arrived
in North Africa
on 8 January
1944, the War Department
had imposed a deadline
of 1 March for the
revision of the
NATOUSA command
structure. Devers'
arrival also roughly
coincided with another
exchange between
SOS, NATOUSA, and
Headquarters,
NATOUSA, about more
men for the burgeoning
supply responsibilities
in the theater.
Within
a week in late January
General Larkin received
two contradictory
orders from NATOUSA.
The first instructed
him to tap the existing
base sections for
manpower, a course
he was reluctant
to take since it
would rob already
shorthanded organizations
in his nominal chain
of command; the
second canceled
the authority to
secure manpower
from even that source
and removed manpower
allocations authority
for base sections
entirely to the
NATOUSA level.
On
14 February Devers
called the conference
that restructured
the theaters. His
NATOUSA General
Order Number 12,
effective 24 February,
transferred all
duties and responsibilities
of COMZ, NATOUSA,
originally set up
only as a rationale
to support the position
of deputy theater
commander, to SOS,
NATOUSA. In the
month after the
meeting the NATOUSA
staff took much
of the theater reduction
in manpower.
While
the staff did not
disappear altogether,
its functions became
almost entirely
identified with
the American side
of AFHQ. Headquarters,
NATOUSA, concerned
itself with matters
of broad policy
at the theater level,
and General Larkin
formally assumed
command of all base
sections in the
theater and the
service and supply
functions between
them and the combat
zones.
Consistent
with this general
transfer and with
a subsequent NATOUSA
staff memorandum,
the AFHQ-NATOUSA
engineer retained
only policy and
planning responsibility.
He could initiate
broad directives,
recommend theater
wide engineer stock
levels, write training
directives and standards,
recommend troop
allocations in the
communications zone,
maintain technical
data on Allied or
enemy engineer equipment
or doctrine, and
provide analyses
of operations plans
and American engineer
commitments in the
theater. The broader
engineer aspects
of Allied military
government also
fell within his
purview.
In
General Larkin's
SOS, NATOUSA,
executive agency,
the SOS engineer
had unfettered jurisdiction
over operational
engineer matters
in the theater COMZ.
He controlled engineer
units assigned to
that command, governed
the issue of nonstandard
equipment to all
American engineer
troops, ruled on
all requests to
exceed accommodation
scales, and handled
all American
real estate questions.
He also controlled
the issue of engineer
supply to Allied
forces, coordinating
with AFHQ only on
British requests.
He was responsible
for taking general
operational directives
emanating from AFHQ
and preparing supply
requisitions and
bills of materials
to support stated
theater programs
and policies.
When
the Fifth Army Base
Section at Naples
became the Peninsular
Base Section (PBS)
on 25 October 1943,
it passed from Fifth
Army control to
the still divided
American theater
command in North
Africa. Until February
1944 the base section
in the Mediterranean
came under NATOUSA
headquarters for
command and administration
but answered to
General Larkin's
SOS, NATOUSA, organization
for supply. General
Pence's PBS command
also had some responsibilities
to the 15th Army
Group in administrative
areas, especially
those affecting
the Italian population.
As
Fifth Army moved
north, base section
jurisdiction grew,
the army rear boundary
was always the PBS
forward boundary.
The base section
engineer, Col. Donald
S. Burns, submitted
his first consolidated
estimates for the
supply requirements
of the Fifth Army
engineers, the III
Air Service Area
Command, and various
other branches of
the PBS Engineer
Service and the
Petroleum Branch
on 15 October 1943,
but the Fifth Army
G-4 continued to
prepare engineer
requisitions until
December, when the
responsibility shifted
entirely to PBS
for Fifth Army and
base section engineer
supply.
Requisitions
then went directly
from PBS to SOS,
NATOUSA, and its
successor command,
designated Communications
Zone, NATOUSA, on
1 October 1944.
Exactly one month
later the theater
command changed
from NATOUSA to
Mediterranean Theater
of Operations (MTOUSA).
On 20 November the
COMZ structure was
eliminated and its
functions passed
to the G-4 and the
special staff of
the MTOUSA head
quarters, which
then handled engineer
requisitions and
other supply for
the theater.
While
the theater reorganization
was bringing order
to the higher echelons
on the American
side of AFHQ and
its immediately
subordinate commands,
several important
changes also occurred
in Fifth Army's
command and administration
of its engineers
and other service
troops. Col. Frank
O. Bowman, the Fifth
Army engineer, promoted
to brigadier general
on 22 February,
became convinced
by early spring
of 1944 of the necessity
of obtaining direct
command of all Fifth
Army engineer troops.
Other technical
service staff officers
shared this idea,
particularly General
Clark's ordnance
officer, Col. Urban
Niblo.
On
26 March 1944, all
corps and army engineer
units were assigned
to a new Fifth Army
Engineer Command.
Corps engineer units,
however, remained
attached to their
respective corps.
Accordingly, though
General Bowman obtained
administrative and
supply control over
all engineer units
except those organic
to divisions, he
did not have operational
control over those
attached to corps.
His headquarters,
designated a major
command of the Fifth
Army, had an operational
and administrative
status similar to
a general staff
division, and he
had the authority
he considered necessary
to meet his responsibilities.
He could move army
engineer troops
from point to point
on his own authority
and could transfer
Fifth Army engineers
from American to
British sectors
and back.
Below
General Bowman in
the Fifth Army engineer
organization were
corps engineer sections,
each with a TOE
calling for only
six officers and
fourteen enlisted
men. Some attempt
was made to obtain
approval for corps
level engineer
commands patterned
after General Bowman's,
but the corps commanders
preferred that the
corps engineer remain
a staff officer
only.
The
engineer combat
regiment was the
mainstay of corps
level engineer strength
at the start of
the Italian campaign,
but in December
1942 War Department
planning revised
the formal and rigid
structure of Army
units, eliminating
the "type army"
and "type corps"
conceptions. The
redivision of forces
that followed placed
engineer units by
functions, under
Army Ground Forces
control if they
supported combat
units or under Army
Service Forces control
if they had primarily
service support
assignments in base
sections or the
communications zone.
Engineer units were
frequently hard
to classify since
the nature of their
assignments and
training carried
them across the
boundaries established
in Army Ground Forces
Commander Lt. Gen.
Lesley J. McNair's
reorganization.
Further
revision of the
unit classification
continued through
1944; at the end
of the year only
divisional engineers
were listed as combat
troops, with nondivisional
engineers supporting
fighting units being
listed as combat
support. At the
same time General
McNair pushed for
economies in service
forces and in staff
overheads in field
commands. He strove
to separate nondivisional
service regiments,
including engineers,
into their component
battalions and to
impose a group headquarters
capable of handling
four battalions
at once in place
of the formal and
traditional regimental
headquarters in
the field.
The
group headquarters
had no units assigned
organically but
controlled the movements
and work assignments
of each battalion
as an attached unit.
In
the summer of 1943,
McNair outlined
his new organizational
precepts in a letter
to all training
commands under his
control. He recommended
that to manage troops
engaged in combat
the higher level
headquarters divide
the administrative
load, making the
corps solely a tactical
headquarters and
limiting field army
headquarters to
overall tactical
supervision with
responsibility for
supply and all other
administrative functions.
The
new program did
make for marked
economies in manpower,
and at the end of
the war the revisions
had contributed
to far more efficient
combat units. But
General McNair's
innovations were
not received with
favor everywhere,
nor were they applied
consistently. The
technical services,
notably the engineers,
had already anticipated
some aspects of
the reform, but
as the distance
from Washington
increased the revision
tended to become
watered down or
compromised with
proven local practice.
Resistance
to the group concept
began at the top
of the Fifth Army
engineer Command
in Italy. When the
War Department authorized
the establishment
of group headquarters
for all service
units in October
1943, the rate of
conversion was left
to the theater command.
General Bowman,
with the concurrence
of General Elliott,
the AFHQ engineer,
slowed down the
adoption of groups,
keeping "the
correspondence about
the change bouncing
between Italy and
Washington."
Bowman believed
that the group organization
hurt morale because
the attachment of
single battalions
to larger units
lasted for only
brief periods. Some
engineer regiments
continued to operate
as such until 1945.
Even
after all the combat
engineer regiments
had converted, arguments
continued over the
value of the change.
General Bowman also
believed that the
various group headquarters
added to administrative
overhead and reduced
even further the
amount of construction
equipment available,
thereby aggravating
an already critical
problem.
The
II Corps engineer,
Col. Leonard B.
Gallagher, held
that the group operated
less efficiently
than the regiment.
Lt. Col. William
P. Jones, Jr., commander
of an engineer battalion
attached to II Corps'
1108th Engineer
Combat Group, contended
that the group wasted
scarce trained engineer
officers and specialists.
There were, however,
strong defenders
of group organization
who stressed the
gain in flexibility
and pointed out
that a group headquarters
could control more
battalions than
could a regimental
headquarters.
The
1108th Combat Group
in 1945, for example,
had under it as
many as seven units
at one time and
for a period supported
five divisions.
The quality of the
group or regimental
commander and the
experience of his
men were the keys
to the effectiveness
of both organizations.
In any case, the
self contained battalion
became a workable
organization.
The
divisional engineers
had both staff and
command responsibilities.
Unlike the G-3,
who thought mainly
in terms of objectives,
a division engineer
was largely concerned
with such matters
as routes of approach,
crossing sites,
and covered assembly
areas for regimental
combat teams in
general did not
have the staff organization
to control
engineer work, so
lost motion and
confusion became
common. The engineers
also maintained
that subordinate
commanders retained
engineer units after
their specific task
was done.
The
nature of engineer
tasks often splintered
engineer units,
regiments, battalions
and detachments
alike. Depot, camouflage,
maintenance, and
dump truck companies
were more susceptible
than others. In
June 1944 the 16th
Armored Engineer
Battalion came together
for the first time
in more than four
months. Such dispersion
inevitably affected
performance, discipline,
and morale, caused
duplication of effort,
and made administration
more difficult.
The
Offensive Resumed
When
the Allied offensive
resumed in May 1944,
the main Fifth Army
line south of Anzio
was to drive north
up the coast to
meet VI Corps troops
breaking out of
the static bridgehead.
North of Anzio,
other VI Corps units
were to strike for
Rome. Preparations
for the renewed
offensive began
in March with a
shift of British
Eighth Army units
westward to take
over the Cassino
and Rapido fronts,
leaving in their
place a garrison
force on the eastern
Italian coast.
Thus
relieved, and with
replacements
arriving to bring
its divisions up
to strength, Fifth
Army consisted of
the American II
Corps and the French
Expeditionary Corps
concentrated on
a thirteen mile
front between the
Italian west coast
and the Liri River,
with II Corps holding
the left flank of
the line. Two fresh
but inexperienced
American divisions,
the 85th and the
88th,
would bear the brunt
of the drive along
Highway 7 to effect
a junction with
the forces at Anzio,
now reinforced to
a strength of 5
1/2 divisions.
A
devastating artillery
bombardment commencing
at 2300 on 11 May
sparked the offensive
on the southern
front, and at dawn
the Mediterranean
Allied Air Forces
rained destruction
on the enemy rear.
The Anzio breakout
began on 23 May,
and on the twenty-fifth
VI Corps was advancing
toward the Alban
Hills.
The
same day, after
II Corps had driven
sixty miles through
the mountains, the
beachhead and the
Fifth Army main
line were linked
for the first time
when men of the
48th Engineer combat
Battalion, II Corps,
shook hands with
the engineers of
the 36th engineer
Combat Regiment,
VI Corps, outside
the demolished village
of Borgo Grappa.
The linkup was part
of the campaign
that smashed the
German Gustav Line
and the less formidable
Hitler Line,
which the enemy
had thrown across
the Liri valley
and the mountain
ranges flanking
it.
The
nature of the terrain
and the scarcity
of roads made the
Fifth Army's offensive
on the southern
front largely mountain
warfare, in which
the experienced
French corps bore
a major share of
the burden. The
only good road available
to Fifth Army, Highway
7, crossed the Garigliano
near its mouth and
followed the coast
to Formia.
From
there it bent northwest
and passed through
mountains to Itri
and Fondi, then
along the coastal
marshes to Terracina,
where it turned
again to the northwest,
proceeding on a
level and nearly
straight course
through the Pontine
marshes to Cisterna.
Beyond Cisterna
the road led toward
Rome by way of Velletri,
skirting the Alban
Hills to the south.
Highway
7 lay at the extreme
left of the line
of advance, but
it was II Corps'
sole supply route.
Apart from this
highway Fifth Army
had the use of two
or three lateral
roads, a few second
and third-class
mountain roads in
the French corps'
area, and some mountain
trails. Insufficient
as the roadnet was,
it was spared the
sort of destruction
that the enemy might
have been able to
visit upon it in
a less hasty withdrawal.
After
the breakout began,
the engineers labored
night and day to
open the roads and
keep them in shape
under the heavy
pounding of military
traffic. At first
the engineers' chief
concerns were to
erect three additional
Class 40 bridges
over the Garigliano,
two for the French
and one for II Corps;
to strengthen to
Class 30 a bridge
in the French Expeditionary
Corps zone; and
to build several
assault bridges
for troops and mules.
Then
engineers began
improving trails
into roads for jeeps,
tanks, and 2 1/2-ton
trucks, often under
artillery fire.
Starting about the
middle of May the
principal engineer
work was clearing
and repairing Highway
7 and a road leading
across the northern
slopes of the Aurunci
Mountains to Pico
on lateral Highway
82.
The
313th Engineer Combat
Battalion, 88th
Division,
undertook swift
construction to
outflank the Formia
corridor on Highway
7. In one day the
men of this battalion
opened a mountain
road that the Germans
had spent two weeks
preparing for demolition.
This road connected
with a trail two
miles long that
the 313th
built in nine hours
over steep hills
that vehicles had
never before traversed.
A few men working
angledozers through
farmland and brick
terraces and along
mountain slopes
did the work. A
German engineer
colonel, captured
a few hours after
the battle and evacuated
over the road, was
amazed, for no road
had been there twenty-four
hours earlier.
At
Itri on Highway
7 a platoon of Company
A of the 310th Engineer
Combat Battalion,
85th Division, built
a 100 foot Bailey
and turned over
its maintenance
to the 19th Engineer
combat Regiment.
The 235th Engineer
Combat Battalion,
a II Corps unit
that normally supported
the 310th, followed
up the 310th's repair
and clearance work
along Highway 7.
The Germans had
destroyed many bridges
between Fondi and
Terracina, and the
American engineers
had to build bypasses
and culverts.
At
a narrow pass between
the mountains and
the sea east of
Terracina,
tank traps and roadblocks,
covered by German
fire from nearby
hills, slowed the
advance along the
highway. When a
blown bridge along
this stretch halted
American tanks,
armored bulldozers
of the 235th and
310th engineer Battalions
and the 19th Engineer
Regiment, all under
fire, built a bypass
that made it possible
to resume the advance.
Lt. Col. Alien F.
Clark, Jr., commanding
the 235th, operated
one of the bulldozers.
When
the advance slowed
at Terracina the
310th Engineer Combat
Battalion immediately
started on an alternate
route to connect
the highway with
Sonnino. A road
capable of carrying
the traffic of an
entire division
had to be cut into
the rocky slopes
of the Ausonia Mountains.
The
engineers' road
building machinery
had done remarkable
things in the mountain
chain during the
drive from the Garigliano,
but this job required
much hand work and
many demolitions,
explosives for which
had to be carried
by hand up rugged
mountain slopes.
The
engineers had cut
six miles of the
new road, with only
one mile left, when
a breakthrough at
Terracina made it
unnecessary to finish
the alternate route.
The work was not
entirely lost, for
the road reduced
the need for pack
mules and made it
possible to move
division artillery
farther forward
to interdict the
road junction at
Sonnino.
Beyond
Terracina the highway
ran thirty miles
straight through
the Pontine
marshes to Cisterna.
All the engineers
available worked
around the clock
repairing and maintaining
three routes through
the marshy flats.
The Germans had
attempted to flood
much of this region
but were only partially
successful;
the water was low
in the streams and
canals.
Nevertheless,
the engineers had
to do considerable
filling along the
main routes as well
as some bypassing
and bridging. When
Highway 7 and the
supplementary routes
were open to the
Anzio beachhead,
troops and supplies
came up from the
southern front in
an uninterrupted
stream. Fifth Army's
momentum was so
great that after
the capture of Rome
on 4 June the advance
proceeded beyond
the city without
pause.
The
Arno
During
the summer advance
to the Arno, about
150 miles, the Fifth
Army front reached
inland approximately
45 miles. Two main
national highways
ran northward in
the army zone. Highway
1 ran northwest
up the coast through
a succession of
important towns,
including Civitavecchia
and Leghorn, to
Pisa, near the mouth
of the Arno. For
most of its length
the highway ran
along a comparatively
flat coastal plain,
nowhere more than
ten miles wide,
but between Cecina
and Leghorn, Highway
1 twisted over mountains
that reached down
to the sea. The
other main road,
Highway 2, wound
through hills, mountains,
and river valleys
along a route that
led from Rome through
Siena to Florence.
There were five
good two-way lateral
roads in the area
between Rome and
the Arno; numerous
smaller roads were,
for the most part,
narrow and unpaved.
During
the advance to the
Arno the army had
to cross only two
rivers of any size,
the Ombrone and
the Cecina, both
at low water. The
port of Leghorn
fell to the 34th
Division, II Corps,
on 19 July. Beyond
Leghorn lay numerous
canals, but engineers
quickly bridged
them. Four days
later the 34th Division
reached Pisa. The
march in the dry
summer weather took
place in clouds
of dust that drew
artillery fire and
choked the troops.
Soldiers wore goggles
over their eyes
and handkerchiefs
across their noses
and mouths.
Some
of the roads, surfaces
ground through by
military traffic,
were six to eight
inches deep in dust.
Sprinkling the roads
with water was the
best way to lay
the dust, but water
tanks were so scarce
that only the most
important roads
could be sprinkled.
Sometimes the engineers
applied calcium
chloride, but it
was also scarce
and its value questionable.
Engineers had some
success with used
oil, but even that
was in short supply.
During
the June and July
drive to the Arno
much of Fifth Army's
forces departed
to prepare for ANVIL,
the invasion of
southern France.
The army lost VI
Corps and the French
Corps. That loss
amounted to seven
full Division s,
and the loss of
separate combat
units amounted to
another division.
The nondivisional
engineer units splitting
away at that time
included the 36th
and 540th Engineer
Combat Regiments,
the 48th Engineer
Combat Battalion,
and the 343d and
344th engineer General
Service Regiments.
On
1 June Fifth Army's
assigned strength
had been approximately
250,000; on 1 August
it was little more
than 150,000. Making
up the losses were
the Japanese-American
442d Regimental
Combat Team (which
arrived in May but
left for France
in late September);
two new and inexperienced
U.S. Army infantry
divisions, the 91st
and 92d; and the
first elements,
about a regimental
combat team, of
the untried Brazilian
Expeditionary Force,
which was to grow
to the size of a
division. In August
General Clark gained
control over the
veteran British
13 Corps consisting
of four divisions.
From
mid July to mid
August Fifth Army
made little forward
progress; it paused
to rest, to build
up supplies, and
to prepare for the
ordeal ahead. The
II and IV Corps
held the 35 mile
sector along the
Arno, IV Corps occupying
the greater part
of the line while
the major portion
of II Corps was
in the rear preparing
for the coming offensive.
The troops received
special instruction
in river crossing
and mountain warfare.
Engineer detachments
gave instruction
in handling footbridges
and boats, in scaling
steep banks with
grappling hooks
and ladders, and
in detecting and
clearing mines.
The
Italian campaign
resumed in earnest
on 24 August with
an Eighth Army attack
on the Adriatic
front. The Fifth
Army crossed the
Arno on 1 September,
and on 9 and 10
September II Corps
launched an offensive
north of Florence.
With 13 Corps beside
it, II Corps battled
through the mountains,
capturing strongpoint
after strong point,
and on the eighteenth
reached the Santerno
valley by way of
Il Giogo Pass.
The
88th Division
outflanked Futa
Pass, key to the
enemy's Gothic Line
defenses, and on
the twenty-second
a battalion of the
91st Division secured
the pass. Fifth
Army had breached
one of the strongest
defense lines the
enemy had constructed
in Italy. The attack
had been well timed,
for the Germans
had diverted part
of their strength
to the Adriatic
front to ward off
an Eighth Army blow.
With
Futa Pass in the
hands of Fifth Army
troops, the way
was clear to send
supplies forward
by way of Highway
65 and to prepare
for an attack northward
to Bologna. Rain,
mud, and many miles
of moun divisional
engineer units had
constantly to maintain
the whole length
of the road, especially
north of Futa Pass,
where the pavement
virtually disappeared.
The main inland
supply route for
IV Corps, Highway
64, running from
Pistoia to Bologna,
carried less traffic
than Highway 65
and therefore remained
in somewhat better
condition.
In
preparation for
winter, the engineers
placed snow fences
and stockpiled sand.
They speeded clearance
after snowfalls
to prevent ice formation
and during thaws
to prevent drainage
problems. Foreseeing
that the greatest
difficulty with
snow would come
in the passes leading
to the Po valley,
AFHQ developed a
plan involving joint
transportation
and engineer operations
to clear the roads.
The
plan included control
posts, road patrols,
and a special communications
system to report
conditions
throughout each
day. The engineer
Section, Fifth Army,
prepared a map that
indicated the areas
where trouble could
be expected, including
areas the Germans
held. The engineer
and transportation
units involved piled
sand along the roads
where the most snow
could be expected
and parked snow
removal equipment
at strategic points
along the roads.
The
plan worked in the
II Corps area, where
winter conditions
were the most severe.
In addition to American
and British troops,
hundreds of Italians,
both civilian and
military, worked
to keep the roads
open. Large rotary
snowplows augmented
jeeps, graders,
bulldozers, and
wooden and conventional
snowplow attachments
fitted to 1/2 and
4 ton trucks. Some
German and Italian
equipment the enemy
had left behind
also proved useful.
Unfortunately,
the plan did not
develop successfully
all along the front.
IV Corps was not
able to set up a
system comparable
to the one II Corps
employed because
IV Corps did not
have anything like
the snow removal
equipment of II
Corps. Instead,
IV Corps units had
to drop whatever
they were doing
when snow began
to fall and clear
the roads with whatever
equipment was available.
Only a few roads
in IV Corps' area
were seriously menaced
by snow, however,
and most lay in
the coastal plain.
During
the fall and winter
the engineers were
able to open mountain
trails. Soft banks
and shoulders gave
way readily before
bulldozers, which
widened roads, provided
turnouts on one
lane sections, and
improved sharp curves
and turns. Huge
quantities of rock
were required to
keep these roads
open to a volume
of traffic never
before contemplated.
The 19th Engineers
used 25,000 cubic
yards of rock to
rebuild a 10 1/2
mile stretch of
secondary road adjacent
to Highway 65 in
the Idice valley.
Keeping
the improved trails
open as roads necessitated
unending work, including
draining, graveling,
revetting soft shoulders,
removing slides,
and building rock
retaining walls.
The greatest problem
was drainage maintenance,
for the mountain
creeks, gullies,
gorges, and cascades,
when not properly
channeled, poured
floods upon the
roads. Two months
of constant work
by thousands of
civilians and soldiers
using both hand
labor and machinery
not only kept the
roads open but improved
them. In forward
areas infantry units
took over the maintenance
of some of the lateral
roads leading to
their dispersed
forces.
The
first of the units
reorganized according
to the new group
concept began operations
in December 1944.
To improve control
over miscellaneous
engineer units operating
under the Fifth
Army engineer, General
Bowman organized
the 1168th Engineer
combat Group, with
Lt. Col. Salvatore
A. Armogida in command.
The cadre for the
new command came
from an antiaircraft
headquarters, and
under it were such
engineer units as
a map detachment,
dump truck companies,
a heavy equipment
company, a maintenance
company, a fire
fighting detachment,
a camouflage company,
a topographic company,
and a water supply
company. Also attached
were some Italian
engineer battalions
and a number of
other units under
an Italian engineer
group.
The
Final Drive
Exceptionally
mild weather beginning
in mid February
enabled engineers
to make substantial
progress in repairing
and rehabilitating
the road nets and
improving and extending
bridges. With snow
rapidly receding
from the highlands,
a company of the
126th Engineer Mountain
Battalion, organic
to the 10th Mountain
Division, built
a 1,700 foot aerial
tramway over Monte
Serrasiccia (located
18 miles northwest
of Pistoia) on 19
February.
Built at an average
slope of 18 to 20
degrees, the tramway
was finished in
ten hours despite
enemy fire. Casualties
could come down
the mountain side
in three minutes
instead of six to
eight hours. The
tramway hauled blood
plasma, barbed wire,
emergency K rations,
water, and ammunition
up the mountain.
Another timesaver
the battalion contributed
was a 2,100-foot
cableway constructed
on 10 March, when
the 10th Mountain
Division was attacking
over rugged terrain.
Supported by two
A-frames and built
in six hours, the
cableway saved a
six mile trip for
ambulances and supply
trucks.
Lt.
Gen. Lucian K. Truscott,
Jr., became commander
of the Fifth Army
in December when
General Clark moved
up to command the
15th Army Group.
Before the spring
offensive began,
the Fifth Army received
reinforcements
of infantry, artillery,
and reserves. Its
divisions were overstrength
and its morale high
as the troops looked
forward to a quick
triumph over the
sagging enemy. The
British 13 Corps
had returned to
Eighth Army, but
Fifth Army's reinforcements
helped balance that
loss.
In
April the two Allied
armies, carefully
guarding the secrecy
of the movement,
went forward into
positions from which
they could strike
a sudden, devastating
blow against the
enemy. The Fifth
Army front was nearly
ninety miles long,
reaching from the
Ligurian Sea to
Monte Grande, ten
miles southeast
of Bologna. The
IV Corps held the
left of this line—indeed,
the greater part
of it—stretching
from the sea and
through the mountains
as far as the Reno
River, a distance
of about seventy
miles. The II Corps
crowded
extending from 130
to 490 yards. Allied
air strikes had
destroyed the permanent
high level and floating
highway bridges.
The Germans maintained
communication across
the river by ferries
and by floating
bridges, many of
which they assembled
from remnants of
permanent floating
bridges after dark
and dismantled before
daylight.
The
engineers knew that
a huge amount of bridging
would be necessary
to cross the Po. Treadway
bridging was in limited
supply. The 25-ton
pontons of the 1554th
Engineer Heavy Ponton
Battalion would be
essential, as would
many floating Baileys,
which Fifth Army could
borrow from the British.
The width of the Po
required storm boats
as well as assault
boats, heavy rafts,
infantry support rafts,
and Quonset barges
assembled from naval
cubical steel pontons
and powered by marine
motors.
Fifth
Army engineers were
confident that they
could build bridges
on piles eighty
feet deep or more
despite the soft
mud of considerable
depth that formed
the Po's bed. Such
piles came from
U.S. engineer forestry
units working in
southern Italy,
and the long trailers
of the 1554th Heavy
Ponton Battalion
brought them to
the front.
On
22 September 1944,
Fifth Army engineers
distributed a special
engineer report
on the Po throughout
the army. The report
consolidated all
available information,
and revised editions
came out from then
until the actual
crossing. The 1168th
Engineer Combat
Group controlled
camouflage, maintenance,
depot, and equipment
units and provided
administrative service
for some engineer
units not under
its operational
control.
The
46th South African
Survey Company carried
its triangulation
net into the Po
valley, while early
in 1945 the 66th
Engineer Topographic
Company issued 1:12,500
photo-mosaic sheets
covering the area
and special 1:10,000
mosaics of possible
crossing sites.
The 1621st Engineer
Model Making Detachment
produced a number
of terrain models
of the Po valley.
Special
river crossing training
concentrated
mainly on II Corps
engineer units,
but close to the
actual crossing
day Fifth Army switched
bridging to IV Corps.
The
engineer units had
thoroughgoing drills,
and a group of II
Corps' combat engineers
got special instruction
in all the assault
and bridging equipment
the army stockpiled
during the winter.
This group was to
operate with the
troops ready to
make the main movement
across the Po, whether
of II or IV Corps.
Fifth Army had estimated
that a floating
Bailey would be
required in both
II Corps and IV
Corps areas; the
1338th Engineer
Combat Group's 169th
Engineer Combat
Battalion was to
build the II Corps
bridge and the 1108th
Engineer Combat
Group's 235th Engineer
Combat Battalion,
the IV Corps bridge.
During
March and April
the 169th Engineer
Combat Battalion
sent several of
its men to the British
Floating Bailey
Bridge School at
Capua, and in April
the entire battalion
moved to a site
on the Arno west
of Pisa for training
in building the
bridges. The 235th
Combat Battalion
got only a few days
of training and
even that for only
part of the battalion.
The
II Corps engineers
diverted to IV Corps
during the crossing
operation included
operators for storm
boats and Quonset
barges, a company
of the 39th Engineer
Combat Group's 404th
Engineer Combat
Battalion to operate
floating equipment,
the 19th Engineer
Combat Group's 401st
Engineer combat
Battalion, and the
1554th Heavy Ponton
Battalion.
During
the morning of the
twenty-third all
II Corps' bridging
that was readily
available, including
an M1 treadway bridge,
60 DUKWs, 4 infantry
support rafts, and
24 storm boats with
motors, moved in
convoy to IV Corps.
At Anzola fifty
assault boats belonging
to IV Corps joined
the convoy, which
went forward to
the 10th Mountain
Division and arrived
at San Benedetto
on the morning of
24 April. On the
night of the twenty-second,
fifty other IV Corps
assault boats had
also reached the
10th Mountain Division.
The
crossing began at
noon on 23 April,
when troops of the
10th Mountain
Division ferried
over the Po in IV
Corps assault boats
operated by Division
al engineers of
the 126th Engineer
Mountain Battalion.
Some of the men
of the 126th made
as many as twenty-three
trips across that
day. Starting at
noon the engineers
used the only equipment
available to them
fifty sixteen-man
wooden assault boats.
By 2000 the 126th
had ferried across
the 86th and 87th
Mountain Infantry
Regiments plus medical
detachments and
two battalions of
divisional light
artillery (75-mm.
pack). Only twelve
boats were left,
most of the rest
having been destroyed
by heavy German
fire. The engineers
suffered twenty-four
casualties, including
two killed.
The
85th Division followed
close behind. All
assault river crossing
equipment the divisional
engineers (the 310th
Engineer Combat
Battalion) had held
had been turned
over to IV Corps
engineers in April
before the Po crossing.
When the division
reached the Po its
engineers had only
nine two man rubber
boats and had to
use local materials
to build four infantry
support rafts and
three improvised
rafts. On these,
with the help of
the 255th Engineer
combat Battalion
of the 1108th Engineer
Combat Group, the
310th crossed all
reconnaissance and
combat units of
the division except
medium artillery.
The crossing took
forty-eight hours,
but in spite of
enemy artillery
fire the engineers
suffered no casualties.
The
IV Corps engineers
had not expected
to be in the vanguard
crossing the Po
and had to cope
with problems for
which they were
not prepared. During
the afternoon of
24 April the 401st
Engineer Combat
Battalion, a II
Corps organization
on loan to IV Corps,
started building
a treadway bridge
near San Benedetto.
Working all night,
with the help of
the 235th Engineer
combat Battalion,
the 401st completed
the 950-foot span
at 1030.
Late
on the afternoon
of 24 April the
1554th Heavy Ponton
Battalion (II Corps)
started work three
miles up-stream
on a heavy ponton
bridge even though
much of the equipment
did not arrive until
after the bridge
had been completed
with improvised
equipment. When
finished on the
afternoon of the
twenty-fifth the
bridge was 840 feet
long and consisted
of 56 pontons, 49
floats, and 4 trestles.
A ferry of Navy
Quonset barges,
which could haul
two 2 1/2-ton trucks,
had operated all
during the night
of 24 April. Day
and night, for forty-eight
hours after the
completion of these
first two bridges
over the Po, two
IV Corps divisions
and part of a II
Corps division went
over the river;
within the first
twenty-four hours
some 3,400 vehicles
crossed the bridges.
Meanwhile,
II Corps' engineers
seriously felt the
diversion of men
and equipment to
IV Corps, which
left them with no
floating bridges
or assault equipment.
Much equipment supposedly
still available
to II Corps was
lost, misplaced,
defective, or still
in crates. During
the night of the
twenty-third bridging
equipment began
to arrive, but treadway
equipage was loaded
on quartermaster
semitrailers instead
of Brockway trucks.
On the morning of
the twenty-fourth
the II Corps engineer,
Col. Joseph O. Killian,
reported to General
Bowman that he had
no bridging available
and that he had
no idea when it
would be available
since treadway construction
depended upon Brockways
with their special
facilities for unloading.
The
Brockways had gone
to IV Corps, and
Colonel
Killian had to depend
upon Fifth Army
engineers for other
equipment.
Also, many motors
for Quonset barges
that reached the
river were defective.
These conditions
held up operations
for almost a day.
The confusion appreciably
reduced II Corps
engineer support
to division engineers
and led to last
minute changes in
plans and hasty
improvisations.
The M2 treadway
and ferries remained
the chief means
for crossing the
Po in the II Corps
area until missing
parts for the Quonsets
arrived from Leghorn.
After
the Po the hard
pressed II Corps
engineers had two
more major streams
to cross, the Adige
and the Brenta,
and again bridging
equipment was late
getting to them.
An almost intact
bridge II Corps
troops seized near
Verona proved sufficient
until other bridges
could be erected.
At the Brenta River
bridging arrived
with the advance
guard of the 91st
Division. One of
the first elements
across a temporary
trestle treadway
at the Brenta was
a section of the
bridge train moving
ahead with forward
elements of the
91st Division to
the next crossing.
In
the IV Corps sector
German defenders
of a bridge across
the Mincio at Governola
held up the forward
drive on the twenty-fourth
only momentarily.
Although damaged,
the bridge proved
usable, and the
37th Engineer Combat
Battalion, which
for more than two
days and nights
had been working
with little or no
rest, had it open
for traffic in a
few hours.
The
drive rolled on,
led by the
88th Division.
The 10th Mountain
Division and the
85th Infantry Division
pushed from
a reserve pool of
knowledgeable men
for new units but
did not replace
them in overseas
units with men of
equal ability. Replacements
in Italy were usually
deficient in engineer
backgrounds, and
some had no technical
knowledge at all.
Between
6 October 1943 and
11 May 1944, forty-eight
officers of company
and field grade
went back to the
United States as
cadre, General Bowman
agreeing that they
could be replaced
by first and second
lieutenants from
training schools
at home. Only some
50 percent arrived
during that period,
and the replacement
system never made
up the shortage.
In the fall of 1944
the War Department
stopped shipping
individual engineer
replacements and
the engineers turned
to hastily trained
elements such as
antiaircraft
gun crews left in
rear areas, usually
ports, to protect
traffic there from
nonexistent Axis
air raids.
From
September 1944 to
April 1945, new
engineer units formed
from nonengineer
organizations included
three combat battalions,
one light equipment
company, one depot
company, one maintenance
company, two engineer
combat group headquarters,
and two general
service regiments.
One general service
regiment and two
combat engineer
regiments already
existing became
group organizations,
and another two
general service
regiments were reorganized
under new tables
of organization
and equipment. But
with the exception
of some separate
companies, none
of the new units
ever attained its
authorized strength.
The constant rotation
of officers to the
United States reduced
some of the existing
units to 85 percent
of their usual strength.
The
number of engineer
units drawn off
by the Seventh Army
in the spring
of 1944 was somewhat
counterbalanced
by the reduction
of Fifth Army's
responsibilities
when the British
Eighth Army took
over a major part
of the front. But
the units lost at
the time were what
remained of the
best, for General
Clark allowed Seventh
Army to take any
engineer unit it
wanted.
Casualties
also took an expected
toll. Of the peak
engineer strength
of 27,000 in June
1944, 3,540 officers
and men were lost.
Of the 831 who died,
597 were killed
in action, 140 died
from wounds received
in action, and 94
died from other
causes. Of the 2,646
wounded in action,
786 were wounded
seriously and 1,860
only slightly. Some
thirty-six were
taken prisoner,
and thirty remained
missing in action.
The
numbers varied from
unit to unit depending
on proximity to
the front line and
the type of work
performed. In forty-five
days of combat at
Anzio, the 36th
Engineer Combat
Regiment lost 74
men killed and 336
wounded. On the
same front, where
it was difficult
to distinguish front
lines from rear,
the 383d Battalion
(Separate) in five
months sustained
casualties of four
officers and eleven
enlisted men killed
and three officers
and fifty-eight
men wounded. Enemy
artillery brought
down the most engineers.
For example, the
109th combat Battalion
between 20 September
1943 and 11 May
1944 had seventyone
battle casualties,
90 percent from
artillery blasts
or shell fragments,
and 10 percent from
mine blasts and
small arms fire.
At other times the
losses from artillery
were fewer, as low
as 61 percent, but
artillery always
remained the chief
culprit.
To
offset inexperience,
the engineers concentrated
on training troops
coming into the
North African theater.
Units had no choice
but to accept troops
without engineer
training, and they
took men with only
basic military training.
They had to be satisfied,
in fact, with only
a small percentage
of Class II personnel
(categorized as
rapid learners in
induction tests),
with the remainder
Class III (average
learners) and Class
IV (slow learners).
New officers were
assigned to four
to six weeks of
duty with rear area
general service
engineer units before
being thrust into
work with combat
engineers.
Each
engineer unit tried
to maintain a reserve
of trained specialists
to fill any vacancies
that occurred and
to keep up job training.
Even so, engineer
units in the Fifth
Army did not have
enough trained operators
and mechanics, especially
for heavy equipment.
A good operator
could do three to
five times the work
of a poor one.
Training
in bridging, river
crossing, mine techniques,
heavy equipment,
motor maintenance,
surveying, intelligence
techniques, mapping,
photography, scouting
and patrolling,
mountain climbing,
driving, marksmanship,
and the use of flame
throwers and grenade
launchers went on
throughout the campaign,
most of it within
the engineer groups,
regiments, battalions,
or companies. Many
units trained at
night. For example,
the 19th Engineer
combat Regiment,
before the spring
offensive of May
1944, spent a third
of its training
time on night practices.
One company of a
battalion might
perform assigned
missions while the
rest of the battalion
trained.
When
the time was available,
almost every unit
practiced bridge
construction. The
235th Engineer Combat
Battalion spent
five days at the
Arno building floating
treadways. Experienced
units trained the
inexperienced: the
16th Armored Engineer
Battalion instructed
the 36th and 39th
Engineer Combat
Regiments and the
10th engineer Combat
Battalion in building
steel treadways,
and the 1755th Treadway
Bridge Company trained
a number of units,
including the 19th
Engineer combat
Regiment. In August
and September 1944
the 175th Engineer
General Service
Regiment conducted
a school for the
British in building
timber bridges.
In April 1944 each
company of the 310th
Engineer Combat
Battalion, 85th
Division, built
and dismantled a
100-foot double-single
Bailey.
As
early as November
1943 Fifth Army
established a school
in river crossing
at Limatola, near
the Volturno, and
here a number of
units practiced
for the Rapido crossing.
During a fortnight
in January 1944
the 16th Armored
Engineer Battalion
practiced assault
crossings with the
6th Armored Infantry
Regiment, 1st Armored
Division. Four companies
of the 19th Engineer
Combat Regiment
practiced between
10 and 15 January
1944 with elements
of the 36th Division
at Pietravairano,
sixteen miles north
of Capua, instructing
the infantry in
the use of river
crossing equipment
during both daylight
and darkness.
The
engineers conducted
similar training
in preparation for
the Arno and Po
crossings.
Engineers
also learned by
attachment.
Units just arriving
in the Fifth Army
zone sent officers
and enlisted men—or
whole units—to
work with, observe,
and learn from engineers
who were more experienced.
Elements of the
310th Engineer Combat
Battalion were attached
to the 313th
Combat Battalion,
elements of the
316th combat Battalion
to the 10th and
111th Combat Battalions,
and elements of
the 48th Combat
Battalion to the
120th Combat Battalion.
The
engineers also instructed
non-engineer units
in a number of other
skills, most notably
recognizing, laying,
detecting, and removing
mines. Two Fifth
Army engineer mine
training teams supplemented
the instruction
that divisional
engineer battalions
gave to the infantry.
The 16th Armored
Engineer Battalion
subjected the 92d
Division to rigorous
drill, requiring
the whole division
to go through a
live mine field.
Early
in the campaign
the British established
a Bailey bridge
school, open to
Americans, at Capua,
where some units
felt the instruction
was better than
that provided at
the American school.46
Americans gave some
supplementary instruction
at the British School
of Military Engineering
at Capua. Most of
the American schools
in the theater were
subordinate to the
Replacement and
Training Command,
MTOUSA. In the summer
of 1944 MTOUSA established
an American Engineer
Mines and Bridge
School along the
Volturno in the
vicinity of Maddaloni.
As the Fifth Army
moved northward
and out of touch,
the school shifted
its emphasis to
converting American
antiaircraft artillery
(AAA) troops into
engineers and to
training the Brazilian
Expeditionary Force
and the 92d Division.
Lacking
engineer troops,
Fifth Army employed
thousands of Italians.
Some Italian engineer
troops participated
in the campaign,
but most of the
laborers were civilians
who bolstered almost
all the U.S. Army
engineer units,
especially those
at army and corps
level. Each unit
recruited its own
civilian force with
help from Allied
military government
detachments. At
one time the 310th
Engineer Combat
Battalion had more
than three times
its own strength
in civilian laborers.
The work of the
Italians, while
not always up to
the standard desired
by the American
engineers, released
thousands of engineers
and infantrymen
for other tasks.
Some
three thousand manual
laborers worked
for the engineers
during the winter
of 1944-45; in April
1945 army and corps
engineer units had
employed 4,437 Italian
civilians, most
of them on road
work. The Italians
loaded, broke, and
spread rock; worked
at quarries; cleared
ditches and culverts
for use of mule
pack trains; and
hand placed rock
to build up firm
shoulders and form
gutters. Those more
skilled rebuilt
retaining walls
and masonry ditches
along road shoulders.
A
specialized Italian
civilian group,
the Cantonieri,
was the equivalent
of
U.S. county or local
road workers. These
workers became available
as the front lines
moved forward and
were especially
valuable in rapidly
moving situations
when engineer road
responsibilities
increased by leaps
and bounds. The
chief of the Cantonieri
of a given area
did the same tasks
on his section of
road (about twelve
miles) that he had
done for his government.
Truckloads of crushed
rock and asphalt
were unloaded along
the road as required,
and the Cantonieri
patched pavements
and did drainage
and other repair
jobs.
Engineer
Supply
Fifth
Army was not in
Italy long before
defects in the engineer
supply system became
evident. The engineers
acted rapidly on
the invasion plans
that called for
them to make the
most use possible
of locally procured
material. Soon after
Naples fell, reconnaissance
parties scoured
the area for supplies,
making detailed
inventories of plumbing
and electrical fixtures,
hardware, nails,
glass, and other
small standard items.
Italian
military stocks,
especially those
at the Fontanelle
caves, were valuable
sources of needed
materiel, and prefabricated
Italian barracks
served as hospital
wards until American
huts arrived. Though
American engineers
sequestered and
classified over
a hundred different
types of stock and
placed orders on
Italian industry
through the Allied
military government
that spurred the
local economy and
saved critical shipping
space, control of
requisition and
issue of supply
suffered from too
few qualified men.
The
strain was particularly
manifest closer
to the combat elements.
No organization
existed at Fifth
Army corps or division
levels to allocate
engineer supply,
and the individual
units drew directly
from army engineer
depots. Though the
Fifth Army engineer
tried to keep the
dumps as far forward
as possible, the
using units had
to send their own
trucks back to collect
supplies since the
depots frequently
did not have the
transportation to
make deliveries.
The time needed
for supply runs
varied with the
distances involved,
the road conditions,
and the frequent
necessity for traveling
blacked out. The
average was one
day, but the 313th
Engineer Combat
Battalion
reported that trips
of up to 250 miles
required two days
for the round trip.
Many
engineer units could
ill afford either
the time or the
transportation required
for frequent trips
back to army dumps,
so they began to
maintain small
dumps of their own,
stocking them with
supplies from army
engineer dumps and
with material captured
or procured locally.
The only condition
Fifth Army imposed
on these dumps was
that all stocks
be movable. It was
common practice
for each company
of a divisional
engineer combat
battalion to set
up a forward dump
in the infantry
regimental sector,
and such dumps often
leapfrogged forward
as the Division
moved. In the 45th
Division, the 120th
Engineer Combat
Battalion in a mobile
situation always
kept its dump about
1 1/2 miles behind
its own command
post.
There
were never enough
depot troops to
operate army engineer
supply dumps.
Before the breakout
in May 1944 Fifth
Army had only one
platoon (one officer
and forty enlisted
men) of the 451st
Engineer Depot Company,
while the rest of
the company remained
with PBS. The platoon
had to move often
to stay close to
the front but still
managed to fill
an average of seventy-five
requisitions every
twenty-four hours.
Frequently, the
platoon operated
more than one depot
simultaneously—three
in May 1944.
When
the 451st concentrated
at Civitavecchia
in June, it took
500 trucks, enough
for seven full-strength
infantry regiments,
to move the unit's
stock and equipment
north. Help in depot
operations came
from other engineers
as well as from
British, French,
and Italian military
units. Several companies
of Italian soldiers
were regularly attached
to the 1st Platoon
as mechanics, welders,
carpenters, and
laborers. Italian
salvage crews repaired
tools and equipment,
manufactured bridge
pins, and mended
rubber boats.
The
shortage of engineer
depot units made
it impossible to
open new engineer
dumps as often or
as rapidly as desirable,
particularly after
the May 1944 breakout.
As a result the
supply furnished
to engineer units
deteriorated, and
in June one platoon
of the 450th Engineer
Depot Company had
to be made available
to Fifth Army. In
August, however,
the platoon reverted
to Seventh Army,
and for the next
few months Fifth
Army again had only
one platoon for
engineer depot support.
Finally,
in December 1944,
MTOUSA formed the
383d Engineer Depot
Company from the
1st Platoon, 451st,
and men from disbanded
antiaircraft units.
Through the rest
of the campaign
Fifth Army engineer
units could count
on supply support
from this company,
aided by Italian
Army troops trained
in engineer supply
procedures.
Mapping
and Intelligence
Planners
had estimated that
Fifth Army would
need a full topographic
battalion, plus
one topographic
company per corps,
to reproduce and
revise maps; yet
there were never
more than two topographical
companies available
at any one time.
The 66th Engineer
Topographic Company
served for nineteen
months; the 661st
served only eight
months, mainly with
VI Corps. Both,
from time to time,
had to get help
from South African
and British survey
companies.
The
66th Topographic
Company was the
American unit on
which Fifth Army
placed its chief
reliance. Upon arrival
in Italy in early
October 1943, the
men of this unit
went to work revising
material derived
chiefly from aerial
photographs. Photo
mosaics and detailed
defense studies
covering the projected
attacks along the
Volturno and Sacco-Liri
Rivers were made
and reproduced.
In
November the 66th
was assigned to
II Corps but continued
to revise and reproduce
maps for the Fifth
Army Engineer Section.
This company consisted
of four platoons:
a headquarters or
service platoon;
a survey platoon,
which as a field
unit performed the
survey
and control work;
a photo mapping
platoon responsible
for drafting as
well as planning
and revising maps;
and a reproduction
platoon responsible
for the lithographic
production of the
printed sheet.
In
January 1944 the
company furnished
men for two provisional
engineer map depot
detachments, one
at Anzio and the
other on the main
front. When the
two fronts merged
in May it was possible
to establish forward
and rear map depots,
and NATOUSA formally
activated the 1710th
and 1712th Engineer
Map Depot Detachments.
The
66th Topographic
Company moved twelve
times between 5
October 1943 and
the fall of Rome
in June 1944. Between
those dates the
company processed
an average of a
half million impressions
a month. In addition
to 866 different
maps, the 66th printed
field orders, overlays
showing engineer
responsibilities,
road network overlays,
defense overprints,
German plans for
Cassino defense,
a monthly history
of II Corps' operations,
the disposition
of German troops
in the II Corps
area, special maps
for the commanding
geeral of II Corps,
special terrain
studies, photomaps,
and various posters
and booklets. It
produced a major
portion of all the
1:100,000, 1:50,000,
and 1:25,000 maps
Fifth Army units
used. In April 1945,
for the Po operation,
the 66th produced
4,900,000 operational
maps, working around
the clock and using
cub planes to speed
distribution to
units.
After
the fall of Rome
the 66th Topographic
Company, then the
only such unit with
Fifth Army, could
not produce the
required amount
of work with its
authorized personnel
and equipment. The
company procured
additional equipment
and employed Italian
technicians and
guards, virtually
becoming a topographic
battalion. Using
the Italian technicians,
the company was
able to work two
shifts reproducing
maps but could not
get enough people
for two shifts on
other jobs. The
company trained
its men for several
different specialties,
but the multiple
responsibilities
overtaxed them.
The
1712th Detachment
issued 1,331,000
maps for the drive
against the Gustav
Line in May 1944.
For the entire Italian
campaign Fifth Army
handled and distributed
over 29,606,000
maps. Ordinarily
the corps maintained
a stock of 500 each
of all 1:25,000
and 1:50,000 sheets
of an area and fewer
1:100,000 and smaller
scale sheets. When
new units arrived
or large orders
came in, the maps
were drawn from
the army map depot;
such orders could
normally be filled
within a day. Periods
of relatively static
warfare in the Italian
campaign called
for large scale
maps. Unfortunately,
not enough 1:25,000
scale maps were
available to meet
the need, and some
of those in stock
were of dubious
quality. The 1:50,000
scale maps provided
complete coverage,
but many panels
were considerably
out-ofdate
and in some cases
illegible.
The
combined sections
of mapping and intelligence
collected data on
weather, crossing
sites, defense works,
observation points,
and fields of fire.
When Lt. Col. William
L. Jones joined
Bowman's staff in
January 1944, intelligence
became divorced
from mapping, and
Jones became chief
of the Plans, Intelligence,
and Training Section.
This arrangement
continued until
September 1944 when
Colonel Jones left
to take command
of the 235th Engineer
combat Battalion;
then mapping and
intelligence reconsolidated
under Lt. Col. John
G. Ladd.
Information
came to the section
from many sources,
including the Army
Map Service and
other agencies in
the United States
and Britain. The
Intelligence Branch,
OCE, WD, supplied
a ten volume work
on Italy's beaches
and ports covering
such subjects as
meteorological conditions
and water supply.
Many studies dealing
with Italy's highway
bridges, railroad
bridges, and tunnels
originated in the
Research Office,
a subdivision of
the Intelligence
Branch. A valuable
source from which
the engineers derived
information was
a sixteevolume Rockefeller
Foundation work
on malaria in Italy
with specific information
concerning the regions
where malaria prevailed.
Lessons, hints,
and tips came from
two series of publications
issued frequently
during the campaign:
Fifth Army Engineer
Notes and AFHQ Intelligence
Summaries.
Although
the Fifth Army G-2
was technically
the agency for collecting
and disseminating
topographic information,
the Fifth Army staff
relied on the engineer
to evaluate all
topographic intelligence
required for planning.
This system worked
well, for by the
nature of his work
and training the
engineer was best
equipped to provide
advice concerning
terrain and communication
routes. Corps and
division staffs
generally expected
less terrain information
from their engineers
because no adequate
photo interpretation
organization existed
below the army level.
engineer intelligence
data seldom covered
terrain more than
one hundred miles
in advance of the
front lines. On
the whole intelligence
was adequate, for
the rate of advance
in Italy was not
rapid enough to
require greater
coverage. The timing
of engineer intelligence
was important; information
conveyed to the
lower units too
far in advance might
be filed away and
forgotten.
Skilled
interpretation of
aerial photographs
was an important
phase of engineer
intelligence. Use
of such photographs,
begun in the stalemate
before Cassino,
proved so valuable
that by February
1944 a squadron
of USAAF P38s
made four to ten
sorties (about 350
pictures) daily.
Two engineers at
the photo center
sent all photographs
within ten miles
of the front forward
and kept the rest
for their own study.
Periodically they
also sent forward
reports on roads,
bridges, streams,
and other features.
The
engineers used long
range terrain reports
of the AFHQ Engineer
Intelligence Section
to plan the forward
movement of engineer
bridge supplies
and the deployment
of engineer units.
The reports were
rich sources of
information on roads
and rivers. Road
information included
width, nature of
surface, embankments,
demolitions, and
suitability for
mules, jeeps or
other transportation.
River information
included bed width,
wet gap, width measured
from the tops of
banks, nature and
height of banks,
levees, potential
crossing places,
approaches, needed
bridging
equipment, fords,
and practicability
of bypasses. The
error was seldom
more than ten feet
for estimated bridge
lengths or 20 percent
for bridge heights.
Sometimes
the terrain reports
were useful in selecting
bombing targets
such as a dam in
the Liri valley.
They could be used
not only to estimate
long range bridging
requirements but
also to anticipate
floods, pinpoint
tank obstacles and
minefields, and
locate potential
main supply routes,
air field sites,
strategic points
for demolition,
and possible traffic
blocks. General
Bowman was so impressed
by the value of
the reports that
he tried repeatedly
to have the AFHQ
Engineer Photo Interpretation
Section made part
of his office, but
AFHQ retained control
of the section.
Camouflage
At
no time during the
entire Italian campaign
were there more
than two companies
of the 84th Engineer
Camouflage Battalion
available, and after
the middle of 1944
only one company
remained with Fifth
Army. Moreover,
since in the United
States camouflage
troops had been
considered noncombatant,
the unit, responsible
for camouflage
supervision and
inspection, consisted
of limited service
and older-thanaverage
personnel. This
policy impaired
efficiency in view
of the fact that
frontline units
had the greatest
need for deception
and disguise. In
addition, the camouflage
companies had neither
enough training
in tactical camouflage
nor enough transportation
to move the large
amount of materials
and equipment required.
In
spite of these handicaps
engineers did some
excellent work with
dummies, paint,
nets, and other
materials. Sometimes
road screens and
dummies confused
and diverted enemy
artillery posted
in the hills. For
example, early in
the campaign, troops
of the 337th engineer
General Service
Regiment erected
a series of structures
made from nine 30-by-30
foot nets, along
a 220 foot stretch
of road near the
Volturno. This section
had been subject
to observation and
shelling, but after
the erection of
the road screen
the shelling stopped.
Road
screens were the
main device in camouflage
operations. As a
rule the engineers
used a double thickness
of garnished net,
but the best type
of net for all purposes
remained an unsettled
question. Engineers
of the 84th Battalion
preferred shrimp
nets to garnished
twine, yet the 15th
Army Group engineer
concluded at the
close of hostilities
that the shrimp
nets had not been
dense enough to
obscure properly.
Pregarnished fish
nets had the same
defect. None of
the nets was sufficiently
durable or fire
resistant. And as
snow fell in December
1944, no white camouflage
materials were available.
The
most ambitious operational
camouflage
programs of the
Italian campaign
took place during
preparations to
attack the Gothic
Line. Engineers
made
every effort to
conceal the II Corps
buildup in the Empoli,
Florence area and
to simulate strength
on the left flank
in IV Corps' Pontedera
sector. Among the
devices employed
were dummy bridges
over canals and
streams and smoke
to make the enemy
believe that heavy
traffic was moving
over the dummy bridges.
One dummy bridge
at a canal southwest
of Pisa drew heavy
fire for two hours.65
In October 1944
in the IV Corps
area, engineers
raised a screen
to enable them to
build a 120 foot
floating treadway
across the Serchio
during the daytime.
During the same
month Company D
of the 84th Camouflage
Battalion erected
a screen 300 feet
long to conceal
all movement across
a ponton bridge
that lay under direct
enemy observation.
The
engineers put up
a forty foot tripod
on each bank of
the river, used
holdfasts to secure
cables, and raised
the screen with
a 3/4-ton weapons
carrier winch and
block and tackle.
In November a bridge
over the Reno River
at Silla, also exposed
to enemy observation,
was screened in
a similar fashion.
Here the engineers
used houses on the
two riverbanks as
holdfasts.
Engineers
set up dummy targets
at bridge sites,
river crossings,
airstrips, and at
various other locations,
building them in
such shapes as artillery
pieces, tanks, bridges,
and aircraft. They
were used to draw
enemy fire to evaluate
its volume and origin.
They also served
to conceal weakness
at certain points,
to permit the withdrawal
of strong elements,
and to conceal buildups.
When a shortage
of dummy material
developed in January
1945, planners looked
upon it as a serious
handicap to tactical
operations.
Dummies
and disguises took
many forms. Large
oil storage tanks
became houses. Company
D used spun glass
to blend corps and
division artillery
with surrounding
snow. The engineers
used painted shelter
halfs and nets with
bleached garlands
to disguise gun
positions, ammunition
pits, parapets,
and other emplacements.
Camouflage proved
valuable enough
in many instances
to indicate that
its wider application
could have resulted
in lower casualties
and easier troop
movements.
Behind
Fifth Army in Italy,
a massive work of
reconstruction continued
as divisions moved
forward against
a slowly retreating
enemy. In the zones
around the major
ports on the western
side of the peninsula
and on the routes
of supply to the
army's rear area,
the base section
made its own contribution
to the war. Suffering
many of the same
strictures and shortages
as Fifth Army engineers,
the Peninsular Base
Section Engineer
Service carried
its own responsibilities,
guaranteeing the
smooth transfer
of men and material
from dockside to
fighting front.
A host of supporting
functions also fell
to the engineer
in the base section,
often taxing strength
and ingenuity to
the same degree
as among the combat
elements.
|