**2. Materials and Methods**

The animal experiments were conducted in accordance with the corresponding German regulations and approved by the Ethics Committee of Lower Saxony for Care and Use of Laboratory Animals (LAVES) (Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit; reference: 33.12-42502-04-15/2044).

#### *2.1. Design of Experiments*

A total of 720 female one-day-old turkeys (B.U.T. Big 6) were obtained from a commercial hatchery (Heidemark GmbH, Ahlhorn, Germany). A total of three independent experiments (T1–T3) were carried out. For each of these experiments, 240 birds were used.

Before starting the experiments in the second week after hatch, the birds were housed in dry and clean floor pens in a quarantine stable. Flooring was covered with wood shavings (GOLDSPAN®, Goldspan GmbH and Co. KG, Goldenstedt, Germany). A commercially prepared pelleted diet was offered ad libitum (Best 3 Geflügelernährung GmbH, Twistringen, Germany).

Each experiment was started after the above described one-week adaptation period. For these experiments, specially manufactured boxes were used, twelve experimental pens (1.20 × 0.80 m) in total. Different flooring designs were used to establish different degrees of contact intensity of the animals to the manure (Figure 1).

**Figure 1.** Flooring designs used in the study: **G1** = entire floor pen with litter; **G2** = identical to G1 and additionally having floor heating (in red); **G3** = plastic covered steel slats in 50% of the pen (in blue) as well as an area with litter; **G4** = fully-slatted flooring with plastic covered steel slats and a sand bath (900 cm2). SB = sand bath,R=rope.

The first group served as a control. Animals were kept on dry wood shavings (G1—entire floor pen covered with litter). The second group was identically kept. The exception was an electrical floor heating system (Sauerland GmbH, Paderborn-Elsen, Germany) with an adjuster to control the temperature (G2—floor pen with litter with floor heating). Animals in these two groups continuously had full contact with manure. The pens in the third group (G3) were divided into two equal parts consisting of 50% solid flooring with wood shavings on the right-hand side and 50% plastic slatted flooring on the left-hand side. In the last group (G4), plastic slatted flooring with a sand bath (900 cm2) was used, the bath being disinfected and the sand replaced on a daily basis. Animals in G4 had no contact with litter except possibly in the sand bath. Plastic covered steel slats consisted of holes (15 × 10 mm) and bridges (plastic covered steel; 3.5 mm wide; Big Dutchman International GmbH, Vechta, Germany). The excreta were stored under the slatted floor at a depth of approximately 30 cm without any material being removed during the trial, besides small amounts of material needed for the samples as described below.

The boxes were placed in a randomized sequence in blocks of four subgroups (G1–G4) in the same stable, as previously described [25]. Two boxes of each block were placed on the right-hand side and two on the left-hand side of a central corridor (~1.70 m width). Airing was provided by vacuum air ventilation. This system was installed in the ceiling in two rows above the pens. Wood shavings were used as bedding material (1 kg/m2). Stocking densities reached a maximum of 25 kg/m2. Hanging type feeders were used (Klaus Gritsteinwerk GmbH & Co. KG, Bünde, Germany) as well as bell drinkers (Ferdinand Stükerjürgen GmbH & Co. KG, Rietberg-Varensell, Germany).

Before commencing with the trials one week after hatch, stables and all materials had been disinfected. Also, tests had been performed to exclude the occurrence of Enterobacteriaceae. All birds were allocated to four groups, each with three identical subgroups (n = 20 birds). Rearing was done until day 36. The birds had unlimited access to fresh water and feed (commercial pelleted growing diet). The environmental temperature was gradually reduced from about 33 ◦C for the one-day-old birds to about 20 ◦C by day 36. Lights were continuously on between days 1 and 3 and the photoperiod from day 4 onwards amounted to 16 h of light and 8 h of darkness.

In T1 there was no antibiotic treatment. This experiment served as a nontreated control trial. In contrast, animals in T2 were medicated with Baytril® 10% in drinking water (10 mg enrofloxacin/kg body weight per day—corresponding to an addition of 0.5 mL Baytril® 10%/L of drinking water, in accordance with the recommended dosage; Bayer Vital GmbH, Leverkusen, Germany). In the last trial (T3), the birds were not treated with any antibiotic in drinking water. Spillage of drinking water containing enrofloxacin was simulated. The amount of water losses was calculated according to experience from former trials, comparing water intake in turkeys using drinking bowls and nipple drinkers (data not shown). Water containing enrofloxacin (dosage: 0.5 mL/L of Baytril® 10%, amount 240 mL per day) was sprayed into the litter or on the slatted flooring only in the feeding area. Both, in T2 and T3, five-day treatments were performed at days 10–14 and days 26–30.

At the end of day 21, eight out of 20 birds in each subgroup were dissected. Final dissection for all remaining turkeys (n = 12/box) was done at day 36. The stunning method (percussive blow to the head) was conducted in accordance with Annex I of the Council Regulation (EC) No. 1099/2009, Chapter I, Methods, Table 1—Mechanical Methods [26].

#### *2.2. Collection of Cloacal Swabs and Manure Samples*

Samples (864 in total) were taken before treatment, directly after antibiotic treatment and at the end of the trial. Cloacal swabs were collected at day 2 and manure samples at day 9 before treatment (BT: before treatment stage). After the enrofloxacin treatment (AT), at day 15, manure samples were taken and six days later, cloacal swabs were taken (day 21). Final sampling (ET) was done for both (manure and cloacal swabs) at day 35. Cloacal swabs were always collected from 24 animals per group, i.e., in total, 96 randomly selected animals. Six samples from each type of flooring design (two samples per pen), in total 24 samples of manure, were taken from two defined locations (feeding area and resting area) in every pen for all trial stages (BT, AT, and ET). Manure samples were taken with a plastic cup (6 cm in diameter) which removed the whole litter material at these locations right down to the floor. All samples were immediately transferred to the laboratory for following analyses.
