Ammonium 2-amino-4-(hydroxymethylphosphinyl)butyrate

Toxicity to terrestrial arthropods

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

toxicity to terrestrial arthropods: long-term

Type of information:

experimental study

Adequacy of study:

key study

Study period:

2017

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

other: OECD Guidance document on honeybee larval toxicity test following repeated exposure, Series on Testing and Assessment N. 239

Version / remarks:

15th July 2016

Deviations:

yes

Remarks:

see below

Qualifier:

according to guideline

Guideline:

other: Regulations (EU) 283/2013 and 284/2013 implementing Regulation (EC) 1107/2009 of 21 October 2009 concerning the placing of plant protection products on the market and repealing Council Directives 79/117/EEC and 91/414/EEC

Principles of method if other than guideline:

Short term deviations from the target test conditions (temperature and relative humidity) were unavoidable due to handling of the test units (e.g. by opening the cabinet door), but did not result in temperatures lower than 23 °C or higher than 40 °C and did not last more than 30 minutes once every 24 hours. Due to possible moulding and avoidance of damaging the larvae/pupae, an additional check has been performed during course of the study, i.e. on day 12. This additional check is not seen to have any impact on the outcome of the study.

GLP compliance:

yes (incl. QA statement)

Application method:

oral

Analytical monitoring:

yes

Details on sampling:

In order to verify the correct preparation of the solutions, duplicate samples of 1000 μL were taken from all stock solutions and freshly prepared feeding solutions on D3, D4, D5 and D6. The concentration of Glufosinate-ammonium in the test samples was determined by high performance liquid chromatography with mass spectrometric detection (LC-MS/MS) using external calibration.

Vehicle:

no

Details on preparation and application of test substrate:

The test item was administered to the larvae via food.

Food prepraration
The food for honey bee larvae was composed of the following three diets, adapted to the needs
of the larvae at different stages of development:

Diet A (D1): 50 % weight of fresh royal jelly + 50 % weight of an aqueous solution containing 2 % weight of yeast extract, 12 % weight of glucose and 12 % weight of fructose.

Diet B (D3): 50 % weight of fresh royal jelly + 50 % weight of an aqueous solution containing 3 % weight of yeast extract, 15 % weight of glucose and 15 % weight of fructose.

Diet C (D4 to D6): 50 % weight of fresh royal jelly + 50 % weight of an aqueous solution containing 4 % weight of yeast extract, 18 % weight of glucose and 18 % weight of fructose.

The aqueous solutions were prepared with ultrapure water1, sterile filtered and frozen after preparation until further use. The fresh royal jelly was provided by Imkerei Bühler, Efringen- Kirchen, Germany and stored in a freezer upon delivery to IES Ltd. It was collected during the preceding 12 months (results of the pesticide screening are shown in Appendix 1). Diet A and Diet B were prepared freshly on D1 and D3, respectively. Diet C was prepared freshly on D4, D5 and D6. All diets were warmed up to 35 °C before use.

Dose preparation
The test item treatments were administrated with the following daily doses. Therefore stock solutions of the test item in ultrapure water for each treatment were prepared
On D3, the larval food was freshly prepared as follow: 3465 mg of Diet B (corresponding to 3150 μL with an assumed density of 1.1 mg/μL) was weighed. Subsequently, 350 μL of the test item or control solution was added accordingly to a final volume of 3500 μL (per treatment).
On D4, the larval food was freshly prepared as follow: 3960 mg of Diet C (corresponding to 3600 μL with an assumed density of 1.1 mg/μL) was weighed. Subsequently, 400 μL of the test item or control solution was added accordingly to a final volume of 4000 μL (per treatment).
On D5, the larval food was freshly prepared as follow: 4455 mg of Diet C (corresponding to 4050 μL with an assumed density of 1.1 mg/μL) was weighed. Subsequently, 450 μL of the test item or control solution was added accordingly to a final volume of 4500 μL (per treatment).
On D6, the larval food was freshly prepared as follow: 4950 mg of Diet C (corresponding to 4500 μL with an assumed density of 1.1 mg/μL) was weighed. Subsequently, 500 μL of the test item or control solution was added accordingly to a final volume of 5000 μL (per treatment).
All the freshly prepared diets were warmed up to 34-35 °C before being administrated to the larvae. Duplicate samples (approx. 1000 μL) of stock solutions as well as feeding solutions were taken from the control and all test item treatments from D3 to D6 for analysis purposes.

Test organisms (species):

Apis mellifera

Animal group:

Hymenoptera (honeybees)

Details on test organisms:

The study was performed with synchronized 1st instar honey bee larvae of the species Apis mellifera L. The honey bees were reared at the facility of IES Ltd. IES colonies are periodically checked and only larvae from healthy, disease-free and queen-right colonies were selected. The colonies used in the study were not treated with any kind of antibiotics or pesticides for at least four weeks before collection of the larvae. To obtain synchronized 1st instar larvae, three days before the start of the test (D-3) the queens of seven different colonies were confined in excluder cages in their own colony. Each excluder cage contained one comb with empty cells in which the queen can lay eggs. Approximately 24 hours after caging the queen (D-2), the presence of freshly laid eggs was verified and the queens were removed from the excluder cages. The combs containing the eggs were left undisturbed in the colonies until hatching. Three days later (D1), the combs containing the freshly hatched 1st instar larvae were removed from the hives and carefully transported to the laboratory.

Study type:

laboratory study

Limit test:

no

Total exposure duration:

4 d

Post exposure observation period:

18 days

Test temperature:

33.4 - 34.5 °C

Humidity:

43.2 - 78.3 %light

Photoperiod and lighting:

No lighting

Details on test conditions:

Test Units
Crystal polystyrene grafting cells with an internal diameter of approximately 9 mm and a depth of 8 mm were used as test units. The cells were sterilized by immersing in Milton® sterilizing fluid (1 % v/v, a.i. sodium hypochlorite 2 % w/w, Laborative Rivadis, Impasse du petit Rosé, ZI 79100 Louzy), for at least 30 minutes and dried in an extractor hood. Each cell was placed into a 48-well tissue cellular culture plate (Cellstar®, Greiner Bio-one, Frickenhausen, Germany) previously half-filled with a piece of wet dental roll. The dental roll was wetted with 15 % (w/v) glycerol, enhanced with Milton® sterilizing fluid (1 % v/v), to maintain the grafting cell at the top level of the plate in a sterile environment. All materials used for the study were sterilized overnight (for approximately 10 hours) using UV-light before usage. Each plate was uniquely identified with study number, treatment and replicate number. The plates were placed into a hermetic acrylic glass desiccator (Nalgene®, Thermo Scientific, Zürich, Switzerland) containing a dish filled with saturated K2SO4 solution in order to keep a relative humidity at a target of 95 ± 5 %, adequate conditions for larval development. The desiccator was placed in a climate cabinet at a target temperature of 34.5 ± 0.5 ºC. On D8, the cells were transferred to a new sterile 48 well-plate without dental rolls and placed into a second hermetic acrylic glass desiccator containing a dish filled with a saturated sodium chloride (NaCl) solution, in order to keep a relative humidity at a target of 80 ± 5 %, adequate conditions for pupal development. The desiccator was placed in a climate cabinet at a target temperature of 34.5 ± 0.5 ºC. For the emergence of pupae, the plates were transferred into emergence boxes on D15. The plates were put in a second climate cabinet at a target temperature of 34.5 ± 0.5 ºC and relative humidity of 65 ± 15 % RH, adequate conditions for emergence. Emerged bees had ad libitum access to 50 % (w/v) sucrose solution provided with a suitable feeder.

Study Design
A total of seven treatments were set up for the study: one control, five test item doses and one reference treatment. Three replicates were established for each treatment, consisting of 16 larvae from three different colonies (1 colony = 1 replicate). All larvae from the same treatment were allocated on the same plate, i.e., a total of seven plates each containing 48 larvae were used for the study. One additional plate, containing 16 larvae from each colony (= replicate), was set up as a reserve. All manipulations (grafting, feeding and mortality assessments) were performed on a warming plate maintained at 35 °C. On D1 a volume of 20 μl of Diet A was dropped into each grafting cell of a 48-well-plate, and each larva was transferred from the comb into a single cell, on the surface of the diet, using a grafting tool. When a plate was filled with 16 larvae from each colony, it was placed in a single layer into a hermetic acrylic glass desiccator incorporating a dish filled with a potassium sulfate (K2SO4) saturated solution.
On D3, day of the first treatment administration, 16 well-developed larvae were selected for each replicate. The larvae from the treatment plates that died due to e.g. mechanical reasons before administering the treated diets B on D3 were replaced by larvae from the reserve plate. Care was taken that larvae from the same colony (=replicate) were replaced. The unused larvae, which were not selected for treatment, were frozen at approximately –20 °C for >12 hours and discarded. Treated diets were freshly prepared and fed according to with a sterilized, disposable pipette tip.
On D4 to D6, the mortality was assessed and dead larvae were removed. The larvae were then fed with freshly prepared treated diets C according to.
On D7, the mortality was assessed and dead larvae were removed.
On D8, the mortality was assessed and dead larvae were removed. In addition, the food consumption was qualitatively assessed. The surviving larvae of each treatment were transferred to a new 48-well-plates which did not contain pieces of dental rolls and were placed in a seond hermetic acrylic glass desiccator incorporating a dish filled with a saturated sodium chloride (NaCl) solution for the pupal stage.
On D15, the mortality was assessed and dead larva and pupae were removed. Each plate was transferred in an emergence box and placed in a climate cabinet at 65 ± 15 % RH. Performance of the test was checked and evaluated on Day 12.
On D22, the emergence of bees was assessed and the study was terminated by freezing all plates at approximately –20 °C for >12 hours.

Experimental Conditions
The mean temperatures in the climate cabinet (D15 – D22) and desiccators (D1 – D8 as well as D8 – D15) were in the range of 33.4 and 34.5 °C throughout the study. The relative humidity was in the range from 54.5 to 97.4 % for the larval (D1 - D8), from 54.2 to 78.3 % for the pupal (D8 - D15) and 43.2 – 65.1 % for the adult emergence period (D15 – D22). Short term deviations were unavoidable due to handling of the test units (e.g. by opening the cabinet door), but did not result in temperatures lower than 23 °C or higher than 40 °C and did not last more than 30 minutes once every 24 hours. Due to a malfunction of the Elpro datalogger, the test conditions (temperature and humidity) of the pupal stage in the climate cabinet (D8 to D15) were erroneously recorded (showing values of zero, i.e. 0.0 °C and 0% HR for the temperature and the humidity, respectively). Since these values don’t reflect reality, they have been excluded.

Evaluations
Larval mortality was assessed at the time of feeding on D4, D5, D6, D7 and on D8 (i.e., 24, 48, 72, 96 and 120 h after the first administration of the test item treatment), respectively. Immobile larvae or larvae that did not react to the contact were noted as dead. Dead larvae were systematically removed and abnormal behaviour was recorded if present. Food unconsumed on D6, D7 and D8 was qualitatively recorded. Pupal mortality was assessed on D15 and D22, emerged adults were counted on D22. Performance of the test was additionally checked and evaluated on Day 12.

Larval mortality
Larval mortality was calculated in percentage by comparing the numbers of bees that died during larval stage, from D3 to D8, to the number of larvae introduced on D3 when dosing started.

Pupal mortality
Pupal mortality was calculated in percentage by comparing the number of pupae that failed to emerge, including those bees without emergence on D22 and dead pupae removed during pupa stage, from D8 to D22, to the number of bees entering the pupal stage on D8. Pupal mortality on D15 was calculated in percentage by comparing the dead pupae removed during pupal stage, from D8 to D15, to the number of bees entering the pupal stage on D8.

Adult emergence rate
Adult emergence rate was calculated in percentage by comparing the number of bees emerged on D22 to the number of larvae on D3 when dosing started.

Nominal and measured concentrations:

Nominal concentrations:
1.0, 2.0, 4.0, 8.0 and 16 μg a.i./larva
resp.
6.5, 13, 26, 52, 104 mg a.i./kg diet

The percentage of the nominal concentrations determined in the feeding solution samples of
diets B and C were within the range of 102 % to 113 %. This confirms the correct preparation
of the feeding solutions.
The concentrations determined in stock solution samples were found to be within the range of
97 % to 110 % of the nominal concentrations. This confirms the correct preparation of the stock
solutions.

Reference substance (positive control):

yes

Remarks:

dimethoate

Duration:

8 d

Dose descriptor:

NOED

Effect conc.:

>= 16 µg per larva

Nominal / measured:

nominal

Conc. based on:

act. ingr.

Basis for effect:

mortality

Remarks:

Larval Mortality D4 - D8

Duration:

15 d

Dose descriptor:

NOED

Effect conc.:

>= 16 µg per larva

Nominal / measured:

nominal

Conc. based on:

act. ingr.

Basis for effect:

mortality

Remarks:

Pupal Mortality D8 - D15

Duration:

22 d

Dose descriptor:

NOED

Effect conc.:

>= 16 µg per larva

Nominal / measured:

nominal

Conc. based on:

act. ingr.

Basis for effect:

mortality

Remarks:

Pupal Mortality D8 - D22

Duration:

22 d

Dose descriptor:

NOED

Effect conc.:

>= 16 µg per larva

Nominal / measured:

nominal

Conc. based on:

act. ingr.

Basis for effect:

emergence

Remarks:

Emergence Rate D3 - D22

Details on results:

Larval mortality and behaviour (D8)
On D8, no mortality occurred in the control. The larval mortality on D8 of the test doses, i.e. 1.0, 2.0, 4.0, 8.0 and 16 μg a.s./larva were 0.0, 2.1, 0.0, 2.1 and 6.3%, respectively. The mortality in the reference item treatment was 92 %, demonstrating the sensitivity of the test system. Larval mortality on D8 was not statistically significant when compared to the control at any of the doses tested, i.e., 1.0, 2.0, 4.0, 8.0 and 16 μg a.s./larva.
No dose-response effect was observed on mortality on D8, therefore LDx and LCx values could not be calculated but determined directly from the Raw Data.
The NOED value for Glufosinate-ammonium on D8 was determined to be ≥16 μg a.s./larva, whereas the LOED value for Glufosinate-ammonium was determined to be >16 μg a.s./larva. The LD10, LD20 and LD50 values on D8 were determined to be >16 μg a.s./larva.
The NOEC value on D8 for Glufosinate-ammonium was determined to be ≥0.11 g a.s./L of diet, whereas the LOEC value on D8 for Glufosinate-ammonium was determined to be >0.11 g a.s./L of diet. The LC10, LC20 and LC50 values on D15 were determined to be >0.11 g a.s./L of diet. No behavioural abnormalities were recorded in the control or any of the test item treatments throughout the study.

Uneaten food
On D8, the food was consumed in all cells containing an alive larva, except for one cell from the control treatment, where uneaten food was observed.

Pupal mortality and behaviour (D15)
On D15, the mortality of honey bee pupae was determined to be 8.3% in the control. The pupal mortality on D15 of the test doses, i.e. 1.0, 2.0, 4.0, 8.0 and 16 μg a.s./larva were 13, 2.2, 6.3, 11 and 7.0 %, respectively. The mortality in the reference item treatment was 100%, demonstrating the sensitivity of the test system. Pupal mortality on D15 was not statistically significant when compared to the control at any of the doses tested, i.e., 1.0, 2.0, 4.0, 8.0 and 16 μg a.s./larva. No dose-response effect was observed on mortality on D15, therefore LDx and LCx values could not be calculated but determined directly from the Raw Data. The NOED value for Glufosinate-ammonium on D15 was determined to be ≥16 μg a.s./larva.
The LOED value for Glufosinate-ammonium was determined to be >16 μg a.s./larva. The LD10, LD20 and LD50 values on D15 were determined to be >16 μg a.s./larva. The NOEC value on D15 for Glufosinate-ammonium was determined to be ≥0.11 g a.s./L of diet.
The LOEC value on D15 for Glufosinate-ammonium was determined to be >0.11 g a.s./L of diet. The LC10, LC20 and LC50 values on D15 were determined to be 1.8, 2.5 and 4.1 g a.s./L of diet. No behavioural abnormalities were recorded in the control or any of the test item treatments throughout the study.

Pupal Mortality and Behaviour (D22)
On D22, mortality of honey bee pupae was determined to be 17 % in the control. The pupal mortality on D22 of the test doses, i.e. 1.0, 2.0, 4.0, 8.0 and 16 μg a.s./larva were 19, 6.4, 35, 11 and 30%, respectively. Pupal mortality on D22 was not statistically significant when compared to the control at any of the doses tested, i.e., 1.0, 2.0, 4.0, 8.0 and 16 μg a.s./larva.
No dose-response effect was observed on mortality on D22, therefore LDx and LCx values could not be calculated but determined directly from the Raw Data.
The NOED value for Glufosinate-ammonium on D22 was determined to be ≥16 μg a.s./larva. The LOED value for Glufosinate-ammonium on D22 was determined to be >16 μg a.s./larva. The LD10, LD20 and LD50 values on D22 were determined to be >16 μg a.s./larva.
The NOEC value on D22 for Glufosinate-ammonium was determined to be ≥0.11 g a.s./L of diet. The LOEC value on D22 for Glufosinate-ammonium was determined to be >0.11 g a.s./L of diet. The LC10, LC20 anC LC50 values on D22 were determined to be >0.11 g a.s./L of diet.
No behavioural abnormalities were recorded in the control or any of the test item treatments throughout the study.

Adult emergence rate (D22)
On D22, the emergence rate of adult honey bees was determined to be 83 % in the control. The adult emergence of the test doses, i.e. 1.0, 2.0, 4.0, 8.0 and 16 μg a.s./larva were 81, 92, 65, 88 and 67 %, respectively. The adult emergence rate in the reference item treatment was 0.0%, demonstrating the sensitivity of the test system. Emergence rate of adult honey bees on D22 was not statistically significant when compared to the control at any of the doses tested, i.e., 1.0, 2.0, 4.0, 8.0 and 16 μg a.s./larva.
No dose-response effect was observed on adult emergence rate on D22, therefore LDx and LCx values could not be calculated but determined directly from the Raw Data.
The NOED value for Glufosinate-ammonium on D22 was determined to be ≥16 μg a.s./larva. The LOED value for Glufosinate-ammonium was determined to be >16 μg a.s./larva. The LD10, LD20 and LD50 values on D22 were determined to be >16 μg a.s./larva.
The NOEC value on D22 for Glufosinate-ammonium was determined to be ≥0.11 g a.s./L of diet. The LOEC value on D22 for Glufosinate-ammonium was determined to be >0.11 g a.s./L of diet. The LC10, LC20 and LC50 values on D15 were determined to be >0.11 g a.s./L of diet.
No behavioural abnormalities were recorded in the control or any of the test item treatments throughout the study.

Analysis of test item concentration in stock and feeding solutions
In none of the control samples were residues at or above the limit of detection (LOD = 0.000003 g active ingredient/L) observed. The percentage of the nominal concentrations determined in the feeding solution samples of diets B and C were within the range of 102 % to 113 %. This confirms the correct preparation of the feeding solutions. The concentrations determined in stock solution samples were found to be within the range of 97 % to 110 % of the nominal concentrations. This confirms the correct preparation of the stock solutions.

Validity criteria
For judging the acceptance and quality of data obtained with the repeated exposure test, the
following criteria were used:
In the control(s), cumulative larval mortality from D3 to D8 should be ≤ 15 % across
replicates; on D8, cumulative larval mortality was determined to be 0.0 % in the control.
In the control(s), the adult emergence rate should be ≥ 70 % on D22; on D22 the adult emergence rate was determined to be 83 % for the control.
In the Reference (positive control), larval mortality should be ≥ 50 % on D8; on D8 cumulative larval mortality for the reference item treatment, i.e., Dimethoate at 7.3 μg a.s./larva corresponding to 0.053 g/L diet or 48 mg/kg diet, was determined to be 92 %.
The study was valid as all the validity criteria were met.

Results with reference substance (positive control):

The adult emergence rate in the reference item treatment of 7.3 µg a.s. / larva was 0.0%,demonstrating the sensitivity of the test system.

Reported statistics and error estimates:

A Dunnett’s Multiple t-test Procedure for mortality on D8, a Step-Down Cochran-Armitage Test Procedure for mortality on D15 and D22 and a Rao-Scott-Cochran-Armitage Test Procedure for adult emergence rate on D22 were performed for the NOED/LOED values (α = 0.05; one-sided smaller). No dose-response effect was observed on mortality (D8, D15 and D22) nor on adult emergence rate (D22), therefore LDx and LCx values could not be calculated and were determined directly from the Raw Data.
The mortality in the treatments was corrected for control mortality by using the formula of Abbott with improvements by Schneider-Orelli.
Statistical analysis was performed using ToxRat Professional Version3.2.1, ToxRat Solutions
GmbH.

Mortality of honey bee larvae on D8

 

Treatment (μg a.s./larva)		Rep.	D4 Number of dead larvae	D5 Number of dead larvae	D6 Number of dead larvae	D7 Number of dead larvae	D8 Number of dead larvae	Mortality D8 (%)	Mean (%)	SD	Abbott1
Control	1	16	0	0	0	0	0	0
	2	16	0	0	0	0	0	0
	3	16	0	0	0	0	0	0	0	0	0
Glufosinate-ammonium
1	1	16	0	0	0	0	0	0
	2	16	0	0	0	0	0	0
	3	16	0	0	0	0	0	0	0	0	0
2	1	16	0	0	0	0	1	6.3
	2	16	0	0	0	0	0	0
	3	16	0	0	0	0	0	0	2.1	2.9	2.1
4	1	16	0	0	0	0	0	0
	2	16	0	0	0	0	0	0
	3	16	0	0	0	0	0	0	0	0	0
8	1	16	0	0	0	0	0	0
	2	16	0	0	0	0	1	6.3
	3	16	0	0	0	0	0	0	2.1	2.9	2.1
16	1	16	0	0	0	0	0	0
	2	16	0	0	2	2	2	13
	3	16	0	0	1	1	1	6.3	6.3	5.1	6.3
Reference (Dimethoate)
7.3	1	16	4	7	13	13	13	81
	2	16	11	11	14	14	15	94
	3	16	10	13	16	16	16	100	92	7.8	92

¹: Mortality corrected with control mortality by using the formula of Abbott with improvements by Schneider-

Orelli.

 

 

Mortality of honey bee pupae on D15

Treatment (μg a.s./larva)		bees entering pre-pupa stage/rep.	No. of dead pupae D15	Mortality D15 (%)	Mean (%)	SD	Abbott1
Control	1	16	0	0
	2	16	4	25
	3	16	0	0	8.3	12	0
Glufosinate-ammonium
1	1	16	1	6.3
	2	16	5	31
	3	16	0	0	13	14	4.5
2	1	15	1	6.7
	2	16	0	0
	3	16	0	0	2.2	3.1	-6.7
4	1	16	2	13
	2	16	1	6.3
	3	16	0	0	6.3	5.1	-2.3
8	1	16	2	13
	2	15	2	13
	3	16	1	6.3	11	3.2	2.6
16	1	16	0	0
	2	14	2	14
	3	15	1	6.7	7	5.8	-1.5
Reference (Dimethoate)
7.3	1	3	3	100
	2	1	1	100
	3	0	n.a.	n.a.	100	0	100

¹: Mortality corrected with control mortality by using the formula of Abbott with improvements by Schneider-

Orelli. A negative value corresponds to a treatment mortality lower than the control mortality.

²: Number of bees still alive on D8 after larval phase.

 

 

Mortality of honey bee pupae on D22

Treatment (μg a.s./larva)		bees entering pre-pupa stage/rep.	No. of dead pupae D22	Mortality D22 (%)	Mean (%)	SD	Abbott1
Control	1	16	0	0
	2	16	8	50
	3	16	0	0	17	24	0
Glufosinate-ammonium
1	1	16	1	6.3
	2	16	7	44
	3	16	1	6.3	19	18	2.5
2	1	15	1	6.7
	2	16	2	13
	3	16	0	0	6.4	5.1	-12
4	1	16	8	50
	2	16	9	56
	3	16	0	0	35	25	23
8	1	16	2	13
	2	15	2	13
	3	16	1	6.3	11	3.2	-7.2
16	1	16	2	13
	2	14	10	71
	3	15	1	6.7	30	29	16
Reference (Dimethoate)
7.3	1	3	3	100
	2	1	1	100
	3	0	n.a.	n.a.	100	0	100

¹: Mortality corrected with control mortality by using the formula of Abbott with improvements by Schneider-

Orelli. A negative value corresponds to a treatment mortality lower than the control mortality.

²: Number of bees still alive on D8 after larval phase.

3: Number of bees that are still in their original cell and that do not show signs of normal development.

 

 

Emergence rate of honey bee adults on D22

Treatment (μg a.s./larva)		larvae/ rep.	No. of NOT hatched bees D22	Emergence D22 (%)	Mean (%)	SD	Abbott1
Control	1	16	0	100
	2	16	8	50
	3	16	0	100	83	29	100
Glufosinate-ammonium
1	1	16	1	94
	2	16	7	56
	3	16	1	94	81	22	98
2	1	16	2	88
	2	16	2	88
	3	16	0	100	92	7.2	110
4	1	16	8	50
	2	16	9	44
	3	16	0	100	65	31	78
8	1	16	2	88
	2	16	3	81
	3	16	1	94	88	6.3	105
16	1	16	2	88
	2	16	12	25
	3	16	2	88	67	36	80
Reference (Dimethoate)
7.3	1	16	16	0
	2	16	16	0
	3	16	16	0	0	0	0

¹: Adult emergence rate corrected with control mortality by using the formula of Abbott with improvements by

Schneider-Orelli. A value >100 corresponds to an adult emergence rate higher than the control adult emergence

rate.

Validity criteria fulfilled:

yes

Conclusions:

The NOEC (resp. NOED) of glyfosinate ammonium was determined as equal or higher then the highest tested concentration (0.11 g a.i. / L resp, 16 µg a.t. / larva).

Reference 2

Endpoint:

toxicity to terrestrial arthropods: short-term

Type of information:

experimental study

Adequacy of study:

key study

Study period:

1999 - 2000

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

comparable to guideline study with acceptable restrictions

Qualifier:

according to guideline

Guideline:

other: SECOFASE, Final Report. Development, improvement and standardization of test systems for assessing sub-lethal effects of chemicals on fauna in the soil ecosystem. (Lokke & van Gestel 1996)

Deviations:

not applicable

Qualifier:

equivalent or similar to guideline

Guideline:

other: Guidance Document for Regulatory Testing Procedures for Pesticides With Non-Target Arthropods (ESCORT 1994)

Deviations:

not applicable

Principles of method if other than guideline:

During exposure, the test units were kept at 51 to 67% RH. The mating vials were kept at 40 to 65% RH and the oviposition vials were kept at 35 to 65% RH. the test animals were kept at dryer circumstances than originally intended. However, the conditions measured in the climate room may not reflect the conditions under which the test animals were kept inside the test units, mating vials and oviposition vials. Conditions there are likely to have been much more moist than conditions outside in the climate room. The performance of the test animals in the water controls indicates that the test animals were in good condition throughout the test.

GLP compliance:

yes (incl. QA statement)

Application method:

spray

Analytical monitoring:

no

Vehicle:

no

Details on preparation and application of test substrate:

The test solutions were applied to the LUFA 2.2 soil contained in the lower half of the glass cages using calibrated laboratory spraying equipment (i.e. a Potter precision spray tower). The Potter Tower was operated at 10 Ibs/in2 (about 0.69 bar) and a table distance of 1.7 cm. Standard settling time was 5 seconds after the sample tube was empty. Spray deposit distribution was documented by spraying water sensitive paper along with the water control treatment: the volume median diameter (VMD) of the droplets was about 200 µm (half the spray volume consisted of droplets smaller, and half of the spray volume consisted of droplets larger than 200 urn) and droplet density was in the range of about 200 to 250 droplets per cm2. The amount of test substance used, 1.75 ml per spray run, was determined during calibration runs at which water was sprayed on a glass plate and the deposit was weighed, immediately preceding the spraying of test substances. The amount used was chosen because it yielded a deposit within 2±0.2 mg of water per cm2, which is equivalent to the desired application volume of 200±20 I/ha, at 5 consecutive spray runs. The test units were sprayed in the following order: test product at increasing concentration, toxic standard and water controls. After application, the application rate was checked again by spraying water and weighing the sprayed glass plate 3 more times. All these 3 spray runs yielded a deposit within 2±0.2 mg of water per cm2.

Test organisms (species):

other: Hypoaspis aculeifer Canestrini (Acari: Laelapidae)

Animal group:

Acari (soil-dwelling predatory mite)

Details on test organisms:

As test species Hypoaspis aculeifer Canestrini (Acari: Laelapidae) has been selected. It is encountered frequently in agricultural sites and can be reared in the laboratory. This species, a gamasid mite living in the soil, is a predator of e.g. bulb mites and nematodes. It was chosen as a sensitive non-target soil dwelling predator, representative for soil micro-faunas. H. aculeifer is a soil mite which can feed on enchytraeids, Collembola, mites and nematodes. It has a potential for regulating plant parasitic nematodes and bulb mites. Adult females can be distinguished from adult males by their much larger size. Adults are darker than juveniles. The strain of H. aculeifer used for this test was reared at the testing facility and was obtained from a rearing of the Department of Population Biology and Systematics from the University of Amsterdam. The latter rearing was initiated with mites collected from stored lily bulbs, harvested in December 1990 in Breezand (North Holland). On 19 March 1999, some individual mites were taken from the UvA rearing and identified as H. aculeifer.

Mites used for the tests and preparing cohorts
A cohort was prepared by putting H. aculeifer females into several jars with a plaster bottom,
darkened with active carbon and supplied with an ample amount of food. This took place 6 days
before exposure. Three days later all females were removed. The eggs laid into the jars yielded the
protonymphs used in the test. Hence the mites used in the test differed maximally 3 days in age.

Food
The MITOX rearing and the cohorts of H. aculeifer are fed with the prey mite Tyrophagus
putrescentiae (Schrank) (Acari: Astigmata), which itself is reared at MITOX on yeast flakes. The
prey rearing was obtained from a rearing on yeast flakes of the Department of Population Biology
and Systematics from the University of Amsterdam. This rearing was initiated with mites on
wheatbran obtained from a commercial supplier, Koppert B.V., Veilingweg 17, P.O. box 155, 2650
AD Berkel en Rodenrijs.
Food was always offered in ample supply to the mites. Food was added at the moment mites were
introduced into the coffin cells. During exposure, food was added to the test units (coffin cells) 7
and 12 days after starting exposure. At transferring surviving female and male mites to mating
vials, 14 days after starting exposure, and also 18 days after starting exposure, food was added to
the mating vials. Food was also added to fresh oviposition vials on each

Study type:

laboratory study

Limit test:

no

Total exposure duration:

28 d

Test temperature:

20.0 – 25.5 °C (For details compare ‘Details on test conditions’)

Humidity:

40 – 67 % (For details compare ‘Details on test conditions’)

Details on test conditions:

Description of the test units
Units in which exposure took place were glass cages ('coffin cells') consisting of a top glass plate (10x5x0.3 cm) with 3 holes (0 = 0.6 cm) and side walls (3 mm high), and a bottom glass plate (10x5x0.15 cm) with side walls (3 mm high). The bottom part of the cells was filled with LUFA 2.2 soil, which was slightly pressed. After this, the bottom parts containing the soil layer of 3 mm were sprayed with the compounds tested. Two-and-a-half to 6 hours after application the cages were assembled by placing the top glass plate and inserting 2 connecting tubes in the outer holes for ventilation purposes. The central hole were covered with a cover slide glass. The holes were used for entering the test mites and for providing food and/or water during the test. Connection parts are described in ref. 4. Two to 4 metal clamps were used to tighten the unit. The test units were not connected to a ventilation system because conditions in the soil would then become too dry for H. aculeifer. Test units for allowing mites to mate and for determining effects on oviposition were small plastic vials (0 = 24 mm, height = 37 mm) with a ±1.5 cm layer of plaster, darkened with active carbon. The same containers were used to store mites temporarily during mortality assessments. The vials were closed with a plastic cap in which a small hole was punctured to prevent over- or underpressure in the vial.

Climatic conditions
All cohorts and all test units (coffin cells, plastic vials) used in this study were kept in the dark in a climate chamber set at 25 °C and 70% relative humidity during all phases of the study. The cohort was kept in a period when temperature in the climate chamber was constantly 25.0 °C and relative humidity ranged from 60 to 65%. Application and ventilating the coffin cells after application took place in the laboratory under ambient conditions. In this period temperature and relative humidity kept within a range of 20 to 21 °C and 40 to 42%, respectively. During the exposure phase temperatures in the climate room kept within a range of 25.0 to 25.3 °C, while relative humidities of 51 to 67% were measured. In the period when female and male mites were kept together in vials to ensure mating, temperature and relative humidity measured were 25.0 to 25.5°C and 40 to 65%, respectively. During the period when female mites were allowed to oviposit temperatures measured in the climate chamber ranged from 25.2 to 25.5 °C and relative humidity measured was 35 to 65%. During exposure, the moist content of the test units (coffin cells) was checked 5, 7 and 12 days after starting exposure. Only 5 and 12 days after starting exposure demineralized water was added because this was considered necessary.

Initiation
On the day exposure was initiated, test units for each treatment were assembled and 20 juvenile mites were entered through one of the holes of the top glass plate of each cell using a fine brush.

Mortality
Mite survival was assessed 14 days after entering the mites into the test units. At this moment the cages were opened and the soil in the cages was thoroughly searched.

Mating
After assessing mortality (14 days after starting exposure), surviving female and male mites were transferred to vials and kept together for 7 days to allow them to mate.

Oviposition and egg hatch success
Seven days after the mortality assessments (i.e. 21 days after starting exposure), 20 impartially selected females per treatment were transferred individually to test units for determining effects on egg production (plastic vials). After 4 days the female mites were transferred to fresh vials. Four days after placing the female mites in the second series of vials the females were removed from them. To assess egg production, eggs and juveniles already hatched were counted on the 2 occasions when female mites were removed from the oviposition vials.

Calculations
Juvenile mortality
Mortality after 14 days of exposure was based on the number of test organisms retrieved alive and the number entered initially as follows: M = (initial- alive) / initial

Test organisms not retrieved were also considered dead. Mortality due to treatment was computed using Abbott's correction

Egg production
Per test unit (i.e. per female), the number of eggs produced in each oviposition period was calculated by summing the number of eggs and juveniles counted for both assessments separately. The total number of eggs produced over the entire 7-day oviposition period was obtained by summing the values obtained for both assessments.

Nominal and measured concentrations:

40 ml/ha, 75 ml/ha, 200 ml/ha, 375 ml/ha, 4000 ml/ha, 7500 ml/ha nominal concentrations

Reference substance (positive control):

yes

Remarks:

dimethoate

Duration:

14 d

Dose descriptor:

NOEC

Effect conc.:

>= 7 500 mL/ha

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

mortality

Duration:

28 d

Dose descriptor:

NOEC

Effect conc.:

>= 7 500 mL/ha

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

reproduction

Details on results:

Performance of the test
Mortality in the water treatment was 23%. In this treatment, the mean total egg production per female over the entire 7-day oviposition period was 25.5 eggs. In the toxic standard treatment mortality was 99%, yielding an Abbott corrected mortality of 99% as well. Mortality in the water controls was compliant the validity criterion of maximally 25%, set in the protocol. Therefore it is concluded that the test organisms were in good condition. The harmful reference indicated that the system is sensitive to pick up harmful effects if these occur.

Juvenile mortality
Mortality due to treatment was 38%, 38%, 5%, 31%, 18% and 9% in the test item 40 ml/ha, 75 ml/ha, 200 ml/ha. 375 ml/ha, 4000 ml/ha and 7500 ml/ha treatments, respectively. In the test item 40 ml/ha, 75 ml/ha and 375 ml/ha treatments, mortality was statistically different from mortality in the water controls. In all other test item treatments mortality was not different from mortality in the water controls, statistically speaking.

Egg production
In the test item 4000 ml/ha treatment a slight and statistically significant reduction in egg production was found during the second observation period. In the test item 7500 ml/ha treatment egg production was not statistically different from egg production in the water controls in the second observation period. No statistically significant overall treatment effects on egg production in the first, second and total oviposition periods were found. Total egg production per female over the entire 7-day oviposition period was 22.4 (88% relative to the control) and 21.9 eggs (86% relative to the control) for the test item 4000 ml/ha and test item 7500 ml/ha treatments, respectively.

Results with reference substance (positive control):

In the toxic standard treatment mortality was 99%, yielding an Abbott corrected mortality of 99% as well. The harmful reference indicated that the system is sensitive to pick up harmful effects if these occur.

Reported statistics and error estimates:

Juvenile mortality
Effects on juvenile mortality were analyzed using Fisher's exact test (ref. 6). Each AE F039866 00 SL14 A116 treatment was tested against the water control separately.

Egg production
Effects on egg production were analyzed using a Kruskal-Wallis test (ref. 6), and a one-way ANOVA followed by Fisher's LSD test (ref. 6), for the first and second oviposition period, respectively. Data of the two oviposition periods combined were analyzed in a Kruskal-Wallis test. ANOVA assumptions, i.e. homogeneity of group variances and normality of residuals, were tested with Bartlett's test and Lilliefors' test, respectively. The results showed that ANOVA assumptions were not met in the first oviposition period (Bartlett's test: P = 0.02 and P = 0.01 for untransformed and square root transformed data, respectively) but were indeed met in the second oviposition period (Bartlett's test: P = 0.69 and Lilliefors test: P = 0.09). Repeated measures ANOVA assumptions were not entirely met in the data combining the two oviposition periods (Bartlett's test: P = 0.02 and P = 0.69 and Lilliefors test: P = 0.36 and P = 0.09 for the first and second oviposition periods, respectively). One outlier in the egg production data of the second oviposition period, was found. Re-running the one-way ANOVA followed by Fisher's LSD test without this outlier did not give rise to qualitatively different results.

Mortality after 14 days exposure.
Treatment ml product/ha	Treatment ml product /L	No. test unit	No. of survivors found	Mortality (proportion)	Mean mortality (%)	Abbott-corrected mortality (%)
Water	-	1	16	0.2	23	-
Water	-	2	19	0.05
Water	-	3	12	0.4
Water	-	4*	7	0.65
Water	-	5	15	0.25
Positive control	-	1	0	1	99	99
Positive control	-	2	0	1
Positive control	-	3	0	1
Positive control	-	4	1	0.95
Positive control	-	5	0	1
7500	37.5	1	12	0.4	30	9 (P=0.335)
7500	37.5	2	15	0.25
7500	37.5	3	15	0.25
375	1.875	1	5	0.75	47	31 (P=0.004)
375	1.875	2	18	0.1
375	1.875	3	9	0.55
75	0.375	1	12	0.4	52	38 (P=0.001)
75	0.375	2	4	0.8
75	0.375	3	13	0.35
4000	20	1	14	0.3	37	18 (P=0.089)
4000	20	2	12	0.4
4000	20	3	12	0.4
200	1	1	18	0.1	27	5 (P=0.691)
200	1	2	13	0.35
200	1	3	13	0.35
40	0.2	1	12	0.4	52	38 (P=0.001)
40	0.2	2	12	0.4
40	0.2	3	5	0.75

Egg production.
	water	4000 ml/ha	7500 ml/ha
1stoeriod(Dav21-25]	(N=19)	(N=18)	(N=20)
mean	13.9	12.6	11.7
st. deviation	2.2	4.5	3.8
reduction to control	NA	0.9	0.84
statistical significance		P=0. 13 (Kruskal-Wallis)	P=0. 13 (Kruskal-Wallis)
2nd oer/od (Dav 25-28)	(N=17)	(N=17)	(N=17)
mean	11.6	9.6	10.8
st. deviation	2.7	2.4	2.9
reduction to control	NA	0.83	0.93
statistical significance		P=0.04	P=0.38

Validity criteria fulfilled:

yes

Conclusions:

Mortality in the water controls was compliant with the validity criterion of maximally 25%, set in the protocol. Therefore it is concluded that the test organisms were in good condition. The harmful reference indicated that the system is sensitive to pick up harmful effects if these occur. In the test item 40 ml/ha, 75 ml/ha and 375 ml/ha treatments, mortality was statistically different from mortality in the water controls. However, in all other test item treatments, including the two highest rates, mortality was not different from mortality in the water controls, statistically speaking. It is believed that the statistical effect found in some rates is due to a high number of escapees in some replicates. Therefore, it is concluded that the statistical effect found in these rates is not biologically significant. In the test item 4000 ml/ha treatment a slight and statistically significant reduction in egg production was found during the second observation period. However, in test item 7500 ml/ha treatment egg production was not statistically different from egg production in the water controls in the second observation period and no statistically significant overall treatment effects on egg production in the first, second and total oviposition periods were found. Therefore, it is concluded that the statistical effect found in the lower rate in the second oviposition period is not biologically significant. No overall effect was calculated because no significant treatment effects on juvenile mortality and overall egg production were observed. In conclusion: the test item, when applied at rates up to 7500 ml/ha does not adversely affect the non-target predatory mite H. aculeifer.

Executive summary:

The test item, when applied at rates up to 7500 ml/ha does not adversely affect the non-target predatory mite H. aculeifer.

Description of key information

For examination of the potential toxicity of glyfosinate ammonium two studies with predatory soil mites (Hypoaspis aculeifer) and the honey bee (Apis mellifera) were performed. Both studies did not reveal statistical significant effects in relevant doses resp. concentrations of glyfosinate ammonium.

The NOEC (resp. NOED) of glyfosinate ammonium was determined as equal or higher then the highest tested concentration (0.11 g a.i. / L resp, 16 µg a.t. / larva). No statistical significant effect regaring larval and pupal mortality nad behavoiur and emergence rate was observed in the study.

Glyfosinate ammonium, when applied at rates up to 7500 ml/ha died not adversely affect juvenile mortality and egg production the non-target predatory mite H. aculeifer.