Target strength (TS) determinations can be classified into several types: 1) theoretical, 2) ex situ measures on dead or living fish under experimental conditions, 3) in situ measures on free swimming fish in their natural habitat. A canvassing of the literature suggested that the type of study may influence the result. For example, some authors suggested that freezing the fish reduce the over-all backscatter of 30% (Sun et al. 1985), others that TS measurements on living fish ex situ would be several dB higher than in situ (Nakken and Olsen, 1977). On other hand the many factors likely to influence TS measurements in situ (swimming movements and aspect, day-night behaviour, physiological state, stomach fullness, sea conditions, trawl type and efficiency) may differ from time to time and place to place (Ona, 1990), making comparisons difficult. For those reasons theoretical models have been thought a necessary support to in situ and ex situ results. Antarctic Silverfish is an important component of Middle Trophic Level of the Ross Sea and several studies have been published on its biology and ecology (De Witt, 1970; De Witt and Hopkins, 1977; Hubold and Hagen, 1997; Vacchi et al., 2004). However, there are no studies concerning TS of P. antarcticum and thus no acoustic assessment of its distribution and abundance in the Ross Sea. In this paper we present results of three different studies on Antarctic Silverfish TS. In situ experiments were conducted on juveniles (length<90mm) on the SE of the Continental Slope (three hauls between Lat. 71°-73° S and Long. 170°-175° E) and of Continental Shelf (one haul around Lat 75° and Long. 167°) during the Italian expedition of January 2004 to the Ross Sea. The mean backscattering cross section of Antarctic Silverfish was estimated comparing the fish density determined acoustically (at 38, 120 and 200 kHz) along the towing track of a pelagic trawl and that determined by the catch in number of fish. The catch was constituted of P. antarcticum for more than 95% in number and weight. Ex situ experiments were conducted on preserved (frozen and defrosted) adults of Antarctic Silverfish, ranging in length from 90 to 210 mm, by means of an underwater framework with a floating platform, moored in Ancona Bay, in May 2007 and in February 2009. The same split beam acoustic system (EK500) as in situ experiments was used, with 38, 120 and 200 kHz transducers bolted to the floating platform over the centre of the frame. The fish were divided in five length classes of eight (2007)/four (2009) individuals: 90-115 mm; 115-130 mm; 130-150 mm; 150-170 mm; 170-210 mm. Single fish were tied by their upper lip and tail to monofilament line stretched between two opposite fibreglass hollow bars of the frame, at a depth of 5-7.4 m from the transducers. The TS was estimated both for only one fish and for all the fish together of each class. We then used a general linear modelling approach to predict TS of Silver fish and its dependence on morphological characteristics of its vertebrae and body.