**2. Materials and methods**

Research was conducted on experimental field located in Zagreb. The soybean variety Zlata (BC Institute Zagreb, Croatia) was planted on April 27th 2010 on an experimental area of 162 m2 . The average plant density was 630.000 plants/ha. Soybean variety Zlata belongs to the maturity group "0" and according to the information given by producers [34] it has a "good" tolerance to pests and diseases. In order to control weeds gyphosate (pre-sowing), metribuzin, metholachlor and clomazone (in the phase of the first trifoliate - V1, according to [35]) and bentazon (in the phase of the third trifoliate - V3) were applied.

Sweep net sampling consisted of making a set of 50 sweeps across three rows of soybeans while walking down the row [36]. A 30 cm diameter sweep net was used. Sampling began when soybeans were in the beginning of flowering (R1) on June 24th 2010 and continued through September 9th 2010 when plants reached physiological maturity (R7). Weekly sam‐ pling was done on the same day each week in late morning. It was performed for 12 weeks. At each sampling date four samples were collected.

Whole plant counts were conducted on 10 plants per each of four replicates. As it was pro‐ posed by Kogan and Pitre [36] randomly selected plants were initially scanned for large, of‐ ten fast moving species. After the initial scan, both sides of each leaf on the plant were searched, as were petioles, axils and stems. Additionally, one leaf per plant was collected at each whole plant count date to establish mite population by leaf inspection. Therefore, four samples each containing 10 leaves were transported to laboratory to be examined under the stereomicroscope and all life stages of mites were counted [37]. Whole plant counts and leaf collection began one week later than sweep net sampling i.e. on July 1st 2010 and continued through September 9th 2010. It was performed for 11 weeks.

All collected insects were identified to the family or genus and species (if possible). For iden‐ tifying insects identification keys were used [38-42].

Based on the number of all individuals, cenological characteristics (dominance and frequen‐ cy) of the insect orders and families (where appropriate) were determined [43].

The dominance was calculated by Balogh formula:

$$\mathbf{D1} = \frac{\mathbf{a}\_1}{\sum \mathbf{a}\_i} \mathbf{x}100^\circ$$

seeks to optimize the use of a combination of methods to manage whole spectrum of pests within particular cropping system. IPM relies heavily on biological controls with a perspec‐ tive chemical input only as a last resort. For effective control, there needs to be an under‐ standing of a pest's interaction with its environment. This is so called concept of "life system" which was initially conceived by Clark et al. [31] to reinforce the idea that popula‐ tion cannot be considered apart from the ecosystem with which it interacts. The life system consists of the pest population plus its "effective environment". Most ecological pest man‐ agement concentrates on the agro-ecosystem, defined as "effective environment" at the crop level [32]. Monitoring in insect pest management can be used to determine the geographical distribution of pests, to assess the effectiveness of control measures, but in its widest sense monitoring is the process of measuring the variables required for the development and use of forecast to predict pest outbreaks [33]. Such forecasts are an important component of pest management strategies because a warning of the timing and extent of pest attack can im‐ prove the efficiency of control measures. For successful pest control according to the princi‐ ples of IPM it is of great importance to have deep knowledge in harmful and beneficial

The study was conducted to determine the harmful and beneficial arthropod fauna during the soybean growing season, and based on their dynamic of occurrence and abundance to identify the harmful and beneficial species of greater importance for soybean production in

Research was conducted on experimental field located in Zagreb. The soybean variety Zlata (BC Institute Zagreb, Croatia) was planted on April 27th 2010 on an experimental area of 162

Sweep net sampling consisted of making a set of 50 sweeps across three rows of soybeans while walking down the row [36]. A 30 cm diameter sweep net was used. Sampling began when soybeans were in the beginning of flowering (R1) on June 24th 2010 and continued through September 9th 2010 when plants reached physiological maturity (R7). Weekly sam‐ pling was done on the same day each week in late morning. It was performed for 12 weeks.

Whole plant counts were conducted on 10 plants per each of four replicates. As it was pro‐ posed by Kogan and Pitre [36] randomly selected plants were initially scanned for large, of‐ ten fast moving species. After the initial scan, both sides of each leaf on the plant were searched, as were petioles, axils and stems. Additionally, one leaf per plant was collected at each whole plant count date to establish mite population by leaf inspection. Therefore, four

to [35]) and bentazon (in the phase of the third trifoliate - V3) were applied.

. The average plant density was 630.000 plants/ha. Soybean variety Zlata belongs to the maturity group "0" and according to the information given by producers [34] it has a "good" tolerance to pests and diseases. In order to control weeds gyphosate (pre-sowing), metribuzin, metholachlor and clomazone (in the phase of the first trifoliate - V1, according

arthoropods in particular agro-ecological conditions.

At each sampling date four samples were collected.

Croatia.

118 Soybean - Pest Resistance

m2

**2. Materials and methods**

Where: a1 = number of identified specimens of one species;

Σa1 = total number of all collected specimens.

The frequency was calculated by Balogh formula:

$$\mathbf{C\_{a1}} = \frac{\mathbf{U\_{a1}}}{\sum \mathbf{U\_i}} \times 100$$

Where: Ua1 = number of samples with identified species;

∑U<sup>i</sup> = total number of samples.
