INSULIN DERIVATIVE FIELD OF THE INVENTION The present invention relates to novel hµman insulin derivatives which are soluble at physiological pH values and have a prolonged profile of action. The invention also relates to methods of providing such derivatives, to pharmaceutical compositions containing them, to methods of treating diabetes and hyperglycaemia using the insulin derivatives of the invention and to the use of such insulin derivatives in the treatment of diabetes and hyperglycaemia. BACKGROUND OF THE INVENTION Currently, the treatment of diabetes, both type 1 diabetes and type 2 diabetes, relies to an increasing extent on the so-called intensive insulin treatment. According to this regimen, the patients are treated with multiple daily insulin injections comprising one or two daily injections of a long acting insulin to cover the basal insulin requirement supplemented by bolus injections of a rapid acting insulin to cover the insulin requirement related to meals. Long acting insulin compositions are well known in the art. Thus, one main type of long acting insulin compositions comprises injectable aqueous suspensions of insulin crystals or amorphous insulin. In these compositions, the insulin compounds utilized typically are protamine insulin, zinc insulin or protamine zinc insulin. Certain drawbacks are associated with the use of insulin suspensions. Thus, in order to secure an accurate dosing, the insulin particles must be suspended homogeneously by gentle shaking before a defined volµme of the suspension is withdrawn from a vial or expelled from a cartridge. Also, for the storage of insulin suspensions, the temperature must be kept within more narrow limits than for insulin solutions in order to avoid lµmp formation or coagulation. While it was earlier believed that protamines were non-immunogenic, it has now turned out that protamines can be immunogenic in man and that their use for medical purposes may lead to formation of antibodies. Also, evidence has been found that the protamine-insulin complex is itself immunogenic. Therefore, with some patients the use of long acting insulin compositions containing protamines must be avoided. Another type of long acting insulin compositions are solutions having a pH value below physiological pH from which the insulin will precipitate because of the rise in the pH value when the solution is injected. A drawback with these solutions is that the particle size distribution of the precipitate formed in the tissue on injection, and thus the release profile of the medication, depends on the blood flow at the injection site and other parameters in a somewhat unpredictable manner. A further drawback is that the solid particles of the insulin may act as a local irritant causing inflammation of the tissue at the site of injection. International patent application published under nµmber WO 2005/012347 (Novo Nordisk A/S) concerns insulin derivatives which have a sidechain attached to either the a-amino group of the N-terminal amino acid residue of the ß-chain or the £-amino group of a Lys residue present in the B chain. International patent application No. EP2006/050593 (Novo Nordisk A/S) discloses insulin derivatives having an aromatic group in the side chain. Patent application no. EP2006/050594 (Novo Nordisk A/S), disclose insulin derivatives having a PEG in the side chain. Other insulin derivatives are disclosed in JP laid-open patent application No. 1-254699 (Kodama Co., Ltd.) and WO 95/07931 (Novo Nordisk A/S). However, there is still a need for insulin having a more prolonged profile of action than the insulin derivatives known up till now. SUMMARY OF THE INVENTION In one aspect of the invention an insulin derivative is having a formula (Formula Removed) wherein Ins is a parent insulin moiety and Q1— Q2—[CH2]n—Xt—[CH2]n —Q3—[CH2]n —X2—[CH2]n —Q4—[CH2]n —X3—[CH2]n —Q5—[CH2]n —Z is a substituent and where the Ins is attached to the substituent via an amide bond between the a-amino group of the N-terminal amino acid residue of the B chain of Ins or an £-amino group of a Lys residue present in the A or B chain of Ins and a CO group in Q1 or Q2 of the substituent; each n is independently 0, 1,2, 3, 4, 5 or 6; Q1is: • an amino acid amide of an amino acid with a carboxylic acid in the side chain, or an amino acid with an uncharged side chain, which residue forms, with its carboxylic acid group, an amide group together with the a-amino group of the N-terminal amino acid residue of the B chain of Ins or together with the E-amino group of a Lys residue present in the A or B chain of Ins, or • a chain composed of two, three or four a-amino acid amide or amino acid residues as specified above linked together via amide bonds, which chain - via an amide bond - is linked to the a-amino group of the N-terminal amino acid residue of the B chain of Ins or to the z- amino group of a Lys residue present in the A or B chain of Ins, or • a bond Q2is: -COCH(CONH2)--COCH2N(CH2CONH2)--COCH2N(CH2CONH2)COCH2N(CH2CONH2) -COCH2CH2N(CH2CH2CONH2)- -COCH2CH2N(CH2CH2CONH2)-COCH2CH2N(CH2CH2CONH2)--COCH2N(CH2CH2CONH2)--COCH2CH2N(CH2CON H2)--COCH2OCH2CONH--CO-((CR5R6)1-6 -NH-CO)=-; -CO-((CR5R6)1-6-CO-NH)1-4-, where R5 independently can be H, —CH3)—(CH2)1. 6CH3 or —CONH2 and R6 independently can be H, — CH3,— (CH2)1-6CH3; or • a bond provided that - at least one of Q1 or Q2 is not a bond, and - that Q2 is not -CO-(CH2)2 -CO-NH- when n is 0 or 1, X1 is a bond and Q3 is (CH2CH2O)2- (CH2CH2O)3-or (CH2CH2OCH2CH2CH2CH2O)- and - that if an amine in Qi or Q2 forms a bond with the rest of the substituent, the amine must be bound to the rest of the substituent via a carbonyl group; Q3, Q4, and Q5 independently of each other can be • -(CH2)m- where m is an integer in the range of 6 to 32; • a divalent hydrocarbon chain comprising 1, 2 or 3 -CH=CH- groups and a nµmber of -CH2- groups sufficient to give a total nµmber of carbon atoms in the chain in the range of 4 to 32; -CO-((CR5R6)1-6 -NH-CO)-; (CO-(CR5R6)1-6CO-NH) 1-4-, where R5 independently can be H, — CH3,—(CH2)1-6CH3 or —CONH2 and R6 independently can be H, — CH3)—(CH2)1-6CH3; • -CO-(CH2) o-3-Ar-(CH2)o-3- where Ar can be arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of -CH3, -(CH)-i^-CH3, -CONR1R2 or -SO2NR1R2, where R1 and R2, independently of each other can be H, -CH3 or - (CH)-i-6-CH3; (CH2CH2O); (CH2CH2CH2O) ; (CH2CH2CH2CH2O) ; (CH2CH2OCH2CH2CH2CH2O)  or (CH2CH2CH2OCH2CH2CH2CH2O); -(CH2OCH2) where y is 1-20; • arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of -CH3, -(CH)i^-CH3, -CONR1R2 or-SO2NR1R2, where R1 and R2, independently of each other can be H, -CH3 or -(CH)i.6-CH3; • a chain of the formula -(CH2),-Yr(Ar)v1 -Y2- (CH2)W-Y3- (Ar)v2-Y4- (CH2)t-Y5- (Ar)v3-Y6- (CH2)Z-wherein Ar is defined as above, Y1 - Y6 independently of each other can be O, S, S=0, SO2 or a bond; where s, w, t and z independently of each other are zero or an integer from 1 to 10 so that the sµm of s, w, t and z is in the range from 4 to 30, and Vi, v2, and v3 independently of each other can be zero or 1 with the proviso that Y1 - Y6do not link to each other and that the structure -0-(CH2)1-0- does not occur; or • a bond; with the proviso that at least one of Q3 - Q5 is not a bond; X1 , X2 and X3 are independently of each other O; -C=0 a bond; NCOR1, where R1 can be H, -CH3 or -(CH)1-6-CH3; or (Formula Removed) where R is hydrogen, C1-3-alkyl, C2_3-alkenyl or C2-3-alkynyl; with the proviso that X1 , X2 and X3 cannot bind to Z and when X^ , X2 and X3 are O, then Xi , X2 and X3 do not bind directly to O in Q3, Q4, and Q5 Z is: -COOH; -CO-Asp; -CO-Glu; -CO-Gly; -CO-Sar; -CH(COOH)2; -N(CH2COOH)2; -SO3H -OSO3H -OPO3H2 -PO3H2 or -tetrazol-5-yl or -O-W1, where W1 is arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of tetrazo-5-lyl, -COOH, -SO3H, -(CH2)1-6-SO3H, -(CH2)i-6—O—PO3H2,-CONR3R4 or-SO2NR3R4, where R3and R4, independently of each other can be H,-(CH2)1-6-SO3H, or -(CH2)1-6—O—PO3H2; provided that when Z is -O-W1 then Q1 must be present and any Zn2+ complex thereof In one aspect of the invention Q2 of the insulin derivative is selected from the group consisting of -COCH(CONH2)- -COCH2N(CH2CONH2)- -COCH2N(CH2CONH2)COCH2N(CH2CONH2) -COCH2CH2N(CH2CH2CONH2)- -COCH2CH2N(CH2CH2CONH2)-COCH2CH2N(CH2CH2CONH2)- -COCH2N(CH2CH2CONH2)- -COCH2CH2N(CH2CONH2)- -COCH2OCH2CONH-; CO-((CR5R6)1-6 -NH-CO)^-; or -CO-((CR5R6)1-6 -CO-NH) ^-, where R5 independently can be H, — CH3l—(CH2)1--6CH3 or —CONH2 and R6 independently can be H, — CH3,— (CH2)1--6CH3 In one aspect of the invention Q3 is -(CH2)m- where m is an integer in the range of 6 to 32 or from 8 to 20 or m is 12, 13, 14, 15 or 16. In one aspect of the invention Q^ Q4, Q5, X1, X2 and X3 is bonds and n is zero. In one aspect of the invention Q3 is -(CH2)m- where m is an integer in the range of 6 to 32 or from 8 to 20 or m is 12, 13, 14, 15 or 16, d, Q4, Q5, X^ X2 and X3 is bonds, n is zeroandZis-COOH. In one aspect of the invention one of Q3, Q4, or Q5 is (CH2CH2O); (CH2CH2CH2O); (CH2CH2CH2CH2O); (CH2CH2OCH2CH2CH2CH2O) or (CH2CH2CH2OCH2CH2CH2CH2O) ; -(CH2OCH2)1-r where y is 1-20. In one aspect of the invention Q3 is -CO-((CR5R6)1--6-NH-CO)-; -(CO-(CR5R6)1--6 -CO-NH) 1-4-, where R5 independently can be H, — CH3,—(CH2)1-. 6CH3 or —CONH2 and R6 independently can be H, —CH3,—(CH2)1-6CH3; • -CO-(CH2) o-3-Ar-(CH2) 0-3- where Ar can be arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of -CH3, -(CH)1-6-CH3, -CONR1R2 or -SO2NR1R2, where R1 and R2, independently of each other can be H, -CH3 or -(CH)^-CH3; or • a bond Q4 is • -(CH2)m- where m is an integer from 4 to 22; • a divalent hydrocarbon chain comprising 1, 2 or 3 -CH=CH- groups and a nµmber of -CH2- groups sufficient to give a total nµmber of carbon atoms in the chain in the range of 4 to 22; • arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of -CH3, -(CH)1-.6-CH3, -CONR1R2 or-SO2NR1R2, where R1 and R2, independently of each other can be H, -CH3 or -(CH)1--6-CH3; or • a chain of the formula -(CH2)s-Yr(Ar)v1 -Y2- (CH2)W-Y3- (Ar)v2-Y4- (CH2),-Y5- (Ar)v3-Y6- (CH2)Z-wherein Ar is defined as above, Y1 - Y6 independently of each other can be O, S, S=O, SO2 or a bond; where s, w, t and z independently of each other are zero or an integer from 1 to 10 so that the sµm of s, w, t and z is in the range from 4 to 30, and v^ v2, and v3 independently of each other can be zero or 1 with the proviso that Y1 - Y6do not link to each other and that the structure -O-(CH2)1--O- does not occur; (Formula Removed) NCOR1, where R1 can be H, -CH3 or-(CH)1-^-CH3; or (Formula Removed) where R is hydrogen, C-|.3-alkyl, C2.3-alkenyl or C2-3-alkynyl; with the proviso that when X1 is O, then X! does not bind directly to O in Q4; X2, X3 and Q5 are bonds; All values of n are zero; and Zis: -COOH; -CO-Asp; -CO-Glu; -CO-Gly; -CO-Sar; -CH(COOH)2; -N(CH2COOH)2; -SO3H -OSO3H -OPO3H2 -PO3H2 or -tetrazol-5-yl or (Formula Removed) where W1 is arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of tetrazo-5-lyl, -COOH, -SO3H, -(CH2)1-6-SO3H, -(CH2)1-4—O—PO3H2,-CONR3R4 or-SO2NR3R4, where R3and R4, independently of each other can be H,-(CH2)1-6-SO3H1 or-(CH2)1-6—O—PO3H2; and any Zn2+ complex thereof. In one aspect of the invention Z is -COOH. In one aspect of the invention the parent insulin of the insulin derivative is an insulin analogue. In one aspect of the invention the parent insulin is selected from the group consisting of: desB30 hµman insulin, GlyA21 hµman insulin, GlyA21desB30 hµman insulin, GlyA21ArgB31ArgB32 hµman insulin, LysB3GluB29 hµman insulin, LysB28ProB29 hµman insulin and ThrB29LysB30 hµman insulin. In one aspect of the invention there is provided a pharmaceutical composition for the treatment of diabetes in a patient in need of such treatment, comprising a therapeutically effective amount of an insulin derivative according to the invention. In one aspect of the invention there is provided a method for producing a pharmaceutical composition comprising a therapeutically effective amount of an insulin derivative according to the invention, wherein up to about 10 zinc atoms per 6 molecules of insulin derivative are added to the pharmaceutical composition. In one aspect of the invention there is provided a method of treating diabetes in a patient in need of such a treatment, comprising administering to the patient a therapeutically effective amount of an insulin derivative according to the invention. In one aspect of the invention the insulin derivative is administered pulmonary. In one aspect of the invention there is provided a mixture of an insulin derivative according to the invention and a rapid acting insulin analogue selected from the group consisting of AspB28 hµman insulin; LysB28ProB29 hµman insulin and LysB3GluB29 hµman insulin. In one aspect of the invention the insulin derivative is selected from the group consisting of: NeB29--carboxpentadecanoyl-L-glutamylamide desB30 hµman insulin, NEB29-carboxpentadecanoyl-amino-butanoyl desB30 hµman insulin, NeB29--carboxtetradecanoyl-L-glutamylamide desB30 hµman insulin, NeB29--carboxtridecanoyl-L-glutamylamide desB30 hµman insulin, NeB29--carboxpentadecanoyl-ß-alanyl desB30 hµman insulin, NeB29--carboxpentadecanoyl-L-aspartylamide desB30 hµman insulin, NeB29--carboxpentadecanoyl--aminohexanoyl desB30 hµman insulin,NeB29--carboxpentadecanoyl--aminopentanoyl desB30 hµman insulin, NeB29-10-(4-carboxyphenoxy)-decanoyl-L-glutamylamide desB30 hµman insulin, NeB29-4-[11-(4-Carboxyphenyl) undecanoylamino] butyryl desB30 hµman insulin, NeB29-(3-(3-{4-[3-(7-carboxyheptanoylaminO)propoxy]butoxy}propylcarbamoyl)-propionyl-glutamylamide) desB30 hµman Insulin, NeB29-- carboxtridecanoyl-amino-butanoyl desB30 hµman insulin, NEB29-carboxundecanoyl- amino-butanoyl desB30 hµman insulin, NeB29--carboxtetradecanoyl-amino-butanoyl desB30 hµman insulin, NeB29-{4-[10-(4-Carboxphenoxy)-decanoylamino]-butyryl} desB30 insulin, NeB29-{4-[(14-CarboxtetradecanoylaminO)-methyl]-benzoyl}desB30 insulin, NeB29-[16-(4-Carboxphenoxy)-hexadecanoyl] desB30 insulin, NeB29-{4-[(15-carboxypentadecanoylaminO)benzoyl]- desB30 hµman insulin and NeB29-{4-[(15-CarboxpentadecanoylaminO)-methyl]-benzoyl}-desB30 insulin DEFINITIONS By "insulin analogue" as used herein is meant a polypeptide which has a molecular structure which formally can be derived from the structure of a naturally occurring insulin, for example that of hµman insulin, by deleting and/or exchanging at least one amino acid residue occurring in the naturally occurring insulin and/or adding at least one amino acid residue. The added and/or exchanged amino acid residues can either be codable amino acid residues or other naturally occurring residues or purely synthetic amino acid residues. In aspects of the invention a maximµm of 17 amino acids have been modified. In aspects of the invention a maximµm of 15 amino acids have been modified. In aspects of the invention a maximµm of 10 amino acids have been modified. In aspects of the invention a maximµm of 8 amino acids have been modified. In aspects of the invention a maximµm of 7 amino acids have been modified. In aspects of the invention a maximµm of 6 amino acids have been modified. In aspects of the invention a maximµm of 5 amino acids have been modified. In aspects of the invention a maximµm of 4 amino acids have been modified. In aspects of the invention a maximµm of 3 amino acids have been modified. In aspects of the invention a maximµm of 2 amino acids have been modified. In aspects of the invention 1 amino acid has been modified. With "desB30 insulin", "desB30 hµman insulin" is meant a natural insulin or an analogue thereof lacking the B30 amino acid residue. Similarly, "desB29desB30 insulin" or desB29desB30 hµman insulin" means a natural insulin or an analogue thereof lacking the B29 and B30 amino acid residues. With "B1", "A1" etc. is meant the amino acid residue at position 1 in the ß-chain of insulin (counted from the N-terminal end) and the amino acid residue at position 1 in the A-chain of insulin (counted from the N-terminal end), respectively. The amino acid residue in a specific position may also be denoted as e.g. PheB1 which means that the amino acid residue at position B1 is a phenylalanine residue. With "insulin" as used herein is meant hµman insulin, porcine insulin or bovine insulin with disulfide bridges between CysA7 and CysB7 and between CysA20 and CysB19 and an internal disulfide bridge between CysA6 and CysA11. By "parent insulin" is meant a naturally occurring insulin such as hµman insulin or porcine insulin. Alternatively, the parent insulin can be an insulin analogue. The expression "uncharged" means that no group or groups that would assµme a charge at pH interval 4 to 9 are present. For example no free carboxylic acids are present. With "fatty difunctionalzed moiety" is meant a carbon chain of 6 to 32 carbon atoms comprising two functional groups selected from carboxy, amino or hydroxyl. The term "non-linking amide" is meant to describe an amide function present in a side chain or side group of a residue present in the substituent, such that said amide bond is not used to connect the residues of the substiuent together. It should be understood that a residue of the substituent in addition to the non-linking amide can comprise further amide groups, eg. amides that bind to other residues of the substituent. "Amino acid amide residue" means the alpha-carboxy amide of an amino acid, or if the amino acid contains a carboxylic acid in the side-chain, "amino acid amide" means amide of either the alpha-carboxy group, or amide of the side-chain carboxy group, as specified. When an insulin derivative according to the invention is stated to be "soluble at physiological pH values" it means that the insulin derivative can be used for preparing insulin compositions that are fully dissolved at physiological pH values. Such favourable solubility may either be due to the inherent properties of the insulin derivative alone or a result of a favourable interaction between the insulin derivative and one or more ingredients contained in the vehicle. The term "no blunting" as used herein means that when formulated in one formulation both the rapid acting insulin and the acylated insulin has profile of action which is identical or substantially identical with the profile of action, when administering the rapid acting insulin and the acylated insulin in separate formulations. The expression "high molecular weight insulin" or "hmw" means that the molecular weight of a complex of hµman insulin, of an insulin analogue or of an insulin derivative is above hµman serµm albµmin, above a dodecameric complex of an insulin analogue or of an insulin derivative or more than about 72 kDalton. The expression "mediµm molecular weight insulin" or "mmw" means that the molecular weight of a complex of hµman insulin, of an insulin analogue or of an insulin derivative is from about an insulin hexamer to about an insulin dodecamer between 24 and 80 kDalton. The expression "low molecular weight insulin" or "Imw" means that the molecular weight of a hµman insulin, an insulin analogue or an insulin derivative is below 24 kDalton. The following abbreviations have been used in the specification and examples: CV colµmn volµme HPLC High Performance Liquid Chromatography HSA hµman serµm albµmin LC liquid chromatography MALDI Matrix Assisted Laser Desorption Ionization MS mass spectrometry RT room temperature SEC size exclusion chromatography SPA Scitillation Proximity Assay Tris tris(hydroxymethyl)aminomethane O.D. optical density = absorbance X2 monomer AspB9 GluB27 hµman insulin DIEA: N,N-diisopropylethylamine DMF: N,N-dimethylformamide Sar: Sarcosine (N-methyl-glycine) tBu: tert-butyl TSTU: 0-(N-succinimidyl)-1,1,3,3-tetramethyluroniµm tetrafluoroborate THF: Tetrahydrofuran EtOAc: Ethyl acetate DIPEA: Diisopropylethylamine TEA: triethyl amine TFA: trifluoracetic acid DCM: dichloromethane RT: room temperature PEG: polyethyleneglycol GIR: Glucose infusion rate All references, including publications, patent applications, and patents, cited herein are hereby incorporated by reference in their entirety and to the same extent as if each reference were individually and specifically indicated to be incorporated by reference and were set forth in its entirety herein (to the maximµm extent permitted by law). All headings and sub-headings are used herein for convenience only and should not be construed as limiting the invention in any way. The use of any and all examples, or exemplary language (e.g., "such as") provided herein, is intended merely to better illµminate the invention and does not pose a limitation on the scope of the invention unless otherwise claimed. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the invention. The citation and incorporation of patent docµments herein is done for convenience only and does not reflect any view of the validity, patentability, and/or enforceability of such patent docµments. This invention includes all modifications and equivalents of the subject matter recited in the claims appended hereto as permitted by applicable law. DESCRIPTION OF THE INVENTION The present invention is based on the recognition that having a substituent in an insulin derivative molecule where only the terminal group is charged, plays an important role for the in vivo duration of action of prolonged-acting insulins, and for the mixability of prolonged-acting insulin with fast-acting insulin with no blunting. Advantageously, insulin derivatives according to the invention are soluble at physiological pH values, have a potency which is comparable to that of hµman insulin, and are mixable with fast-acting insulins with no blunting. The individual profiles of action of mixed basal and bolus insulins are retained in formulations containing Zn(ll) concentrations of up to or less than approximately 3 Zn(ll) per insulin hexamer which limits the risk of precipitations in the formulation, compared to formulations containing more than 3 Zn (II) per insulin hexamer. The invention is sµmmarized in the following paragraphs: An insulin derivative having a formula (Formula Removed) wherein Ins is a parent insulin moiety and Q1—Q2—[CH2]n—X-i—[CH2]n —Q3—[CH2]n —X2—[CH2]n — Q4—[CH2]n — X3—[CH2]n —Q5—[CH2]n —Z is a substituent and where the Ins is attached to the substituent via an amide bond between the a-amino group of the N-terminal amino acid residue of the B chain of Ins or an e-amino group of a Lys residue present in the A or B chain of Ins and a CO group in Q1 or Q2 of the substituent; each n is independently 0, 1,2, 3, 4, 5 or 6; Q1 is: • an amino acid amide of an amino acid with a carboxylic acid in the side chain, or an amino acid with an uncharged side chain, which residue forms, with its carboxylic acid group, an amide group together with the a-amino group of the N-terminal amino acid residue of the B chain of Ins or together with the £-amino group of a Lys residue present in the A or B chain of Ins, or • a chain composed of two, three or four a-amino acid amide or amino acid residues as specified above linked together via amide bonds, which chain - via an amide bond - is linked to the a-amino group of the N-terminal amino acid residue of the B chain of Ins or to the £-amino group of a Lys residue present in the A or B chain of Ins, or • a bond Q2is: -COCH(CONH2)- -COCH2N(CH2CONH2)- -COCH2N(CH2CONH2)COCH2N(CH2CONH2) -COCH2CH2N(CH2CH2CONH2)- -COCH2CH2N(CH2CH2CONH2)-COCH2CH2N(CH2CH2CONH2)- -COCH2N(CH2CH2CONH2)- -COCH2CH2N(CH2CONH2)- -COCH2OCH2CONH- CO-((CR5R6)1-6H-CO)1-4; -CO-((CR5R6)1-6O-NH)1-4where R5 independently can be H, — CH3l—(CH2)1--6CH3 or —CONH2 and R6 independently can be H, — CH3l— (CH2)1--6CH3; or • a bond provided that at least one of Q1 or Q2 is not a bond, and - that Q2 is not -CO-(CH2)2 -CO-NH- when n is 0 or 1, X1is a bond and Q3 is (CH2CH2O)2-, (CH2CH2O)3- or (CH2CH2OCH2CH2CH2CH2O)- and - that if an amine in Q1 or Q2 forms a bond with the rest of the substituent, the amine must be bound to the rest of the substituent via a carbonyl group; Q3, Q4, and Q5 independently of each other can be • -(CH2)m- where m is an integer in the range of 6 to 32; • a divalent hydrocarbon chain comprising 1, 2 or 3 -CH=CH- groups and a nµmber of -CH2- groups sufficient to give a total nµmber of carbon atoms in the chain in the range of 4 to 32; -CO-((CR5R6)1-6-CO)-; -(CO-(CR5R6)1-6-CO-NH) 1-t- where R5 independently can be H, —CH3l—(CH2)1-. 6CH3 or —CONH2 and R6 independently can be H, —CH3l—(CH2)1--6CH3; • -CO-(CH2) 0.3-Ar-(CH2) 0.3- where Ar can be arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of -CH3, -(CH)1-6-CHs, -CONR1R2 or -SO2NR1R2, where R1 and R2, independently of each other can be H, -CH3 or - (CH)1-_6-CH3; (CH2CH2O); (CH2CH2CH2O); (CH2CH2CH2CH2O); (CH2CH2OCH2CH2CH2CH2O)  or (CH2CH2CH2OCH2CH2CH2CH2O); -(CH2OCH2) where y is 1-20; • arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of -CH3, -(CH)1-6-CH3, -CONR1R2 or-SO2NR1R2, where R1 and R2, independently of each other can be H, -CH3 or -(CH)1-.6-CH3; or • a chain of the formula -(CH2)s-Y1-(Ar)v1 -Y2- (CH2)W-Y3- (Ar)v2-Y4- (CH2)t-Y5- (Ar)v3-Y6- (CH2)Z-wherein Ar is defined as above, Y1 - Y6 independently of each other can be O, S, S=O, SO2 or a bond; where s, w, t and z independently of each other are zero or an integer from 1 to 10 so that the sµm of s, w, t and z is in the range from 4 to 30, and v1, v2, and v3 independently of each other can be zero or 1 with the proviso that Y1 - Y6do not link to each other and that the structure -O-(CH2)1--O- does not occur; or • a bond; with the proviso that at least one of Q3 - Q5 is not a bond; X1 , X2 and X3 are independently of each other O; -C=0 a bond; NCOR1, where R1 can be H, -CH3 or -(CH)1-.6-CH3; or (Formula Removed) where R is hydrogen, Q1-alkyl, C2-3-alkenyl or C2.3-alkynyl; with the proviso that X1 , X2 and X3cannot bind to Z and when X1 , X2 and X3are O, then X1 , X2 and X3 do not bind directly to O in Q3, Q4, and Q5 and Zis: -COOH; -CO-Asp; -CO-Glu; -CO-Gly; -CO-Sar; -CH(COOH)2; -N(CH2COOH)2; -SO3H -OSO3H -OPO3H2 -PO3H2 or -tetrazol-5-yl or -O-Wi, where \Ni is arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of tetrazo-5-lyl, -COOH, -SO3H, -(CH2)1--6-SO3H, -(CH2)1--6—O—PO3H2,-CONR3R4 or-SO2NR3R4, where R3and R4, independently of each other can be H,-(CH2)1--6-SO3H, or -(CH2).,_6—O—PO3H2; provided that when Z is -tert-butyl-carboxpentadecanoyl-L-glutamyl amide, 85 mg (80 %). 1H-NMR (CDCI3) 5: 6.98 (s, 1H), 6.60 (d, 1H), 5.88 (s, 1H), 4.69 (m, 1H), 2.55-2.41 (m, 2H), 2.25-2.18 (m, 2H), 2.14 (m, 1H), 1.93 (m, 1H), 1.65-1.54 (m, 4H) 1.44 (s, 9H), 1.27 (br, 20H). Step 4: ro-tert-butvl-carboxv-pentadecanovl-L-glutamvlamide succinimidvl ester co-tert-butyl-carboxpentadecanoyl-L-glutamylamide (85 g, 0.181 mmol) was dissolved in THF (1 ml) and treated with TSTU (65 g, 0.217 mmol) and DIEA (37 jil, 0.217 mmol) and stirred overnight. The mixture was filtered, and the filtrate was evaporated in vacuo. The residue was dissolved in AcOEt and washed twice with cold 0.1 M HCI and water. Drying over MgSO4 and evaporation in vacuo gave -tert-butyl-carboxpentadecanoyl-L-glutamyl amide succinimidyl ester, 91 mg (89%). 1H-NMR (CDCI3) 5: 6.59 (s, 1H), 6.41 (d, 1H), 5.56 (s, 1H), 4.62 (m, 1H), 3.02-2.94 (dd, 2H), 2.84 (s, 4H), 2.71-2.58 (m, 2H), 1.76 (m, 1H), 1.53-1.63 (m, 4H), 1.44 (s, 9H), 1.25 (br, 20H). Step 5: NeB29--carboxv-pentadecanovl-v-L-glutamvlamide desB30 human insulin DesB30 human insulin (500 mg, 0.088 mmol) was dissolved in 100 mM Na2C03 (5 ml, pH 10.2) at room temperature. oo-Tert-butyl-carboxpentadecanoyl-L-glutamyl amide succinimidyl ester (57 mg, 0.105 mmol), was dissolved in acetonitrile (5 ml) and subsequently added to the insulin solution. After 30 mins, 0.2 M methylamine (0.5 ml) was added. pH was adjusted by HCI to 5.5, and the isoelectric precipitate was collected by centrifugation and dried in vacuo to give 423 mg. The coupling yield was 42 % (Rß-HPLC, C4 column; Buffer A: 10 % MeCN in 0.1 % TFA-water, Buffer B: 80 % MeCN in 0.1 % TFA-water; gradient 20% to 90 % B in 16 minutes). The protected product was dissolved in 95 % TFA (12 ml), left 30 mins, and evaporated in vacuo. The crude product was dissolved in water and lyophilized. N£B29--carboxpentadecanoyl-L-glutamylamide desB30 human insulin was purified by Rß-HPLC on C4-column, buffer A: 20 % EtOH + 0.1 % TFA, buffer B: 80 % EtOH + 0.1 % TFA; gradient 15-60 % B, followed by HPLC on C4-column, buffer A: 10 mM Tris + 15 mM ammonium sulphate in 20 % EtOH, pH 7.3, buffer B: 80 % EtOH, gradient 15-60 % B. The collected fractions were desalted on Seß-ßak with 70% acetonitrile + 0.1 % TFA, neutralized by addition of ammonia and freeze-dried. The unoptimized yield was 50 mg, 12 %. The purity as evaluated by HPLC was >98 %. LCMS 6102.8; C274H4i2N66O8oS6 requires 6103.1. Example 2 Synthesis of NEB29-w-carboxv-pentadecanovl-'amino-butanovl desB30 human insulin (Formula Removed) This compound was prepared from hexadecandioic acid and aminobutyric acid, in analogy with example 1. CD-tert-butvl-carboxv-pentadecanoyl-Y-amino-butvric acid succinimidyl ester 1H-NMR (CDCI3) 5: 5.80 (m, 1H), 3.36 (dd, 2H), 2.84 (s, 4H), 2.65 (t, 2H), 2.21-2.13 (m, 4H), 1.99 (p, 2H), 1.44 (s, 9H), 1.66-1.51 (m, 6H), 1.25 (br, 20H). ,eB29 Ne -ro-carboxv-pentadecanoyl-Y-amino-butanovl desB30 human insulin LCMS 6059.9; C273H411N65O79S6 requires 6060.1. Example 3 Synthesis of NeB29-w-carboxv-tetradecanovl-L-qlutamvlamide desB30 human insulin DesB29, DesB30 human insulin —N H (Formula Removed) This compound was prepared from pentadecandioic acid and L-glutamylamide in analogy with example 1. LCMS 6088.2; C273H4ioN66O8oS6 requires 6089.1. Example 4 ieB29 Synthesis of N -co-carboxv-tridecanovl-L-qlutamvlamide desB30 human insulin (Formula Removed) DesB29, DesB30 human insulin —N This compound was prepared from tetradecandioic acid and L-glutamylamide in analogy with example 1. LCMS 6075.3; C272H4o8N66O8oS6 requires 6075.1. Example 5 Synthesis of NEB29-w-carboxv-pentadecanoyl-ß-alanvl desB30 human insulin OH (Formula Removed) DesB29, DesB30 human insulin —N H O This compound was prepared from hexadecandioic acid and (3-alanine in analogy with example 1. LCMS 6044.8; C272H409N65O79S6 requires 6046.1. Example 6 ieB29 Synthesis of N -w-carboxv-pentadecanovl-Y-L-aspartvlamide desB30 human insulin (Formula Removed) This compound was prepared from hexadecandioic acid and L-aspartylamide in analogy with example 1. LCMS 6088.8; C273H410N66O80S6 requires 6089.1. Example 7 Synthesis of N£B29-w-carboxv-pentadecanovl-e-aminohexanoyl desB30 human insulin (Formula Removed) This compound was prepared from hexadecandioic acid and e-amino-hexanoic acid in analogy with example 1. LCMS 6086.1; C275H415N65O79S6 requires 6088.1. Example 8 Synthesis of NeB29-w-carboxv-pentadecanovl--aminopentanovl desB30 human insulin (Formula Removed) This compound was prepared from hexadecandioic acid and 8-amino-pentanoic acid in analogy with example 1. LCMS 6074.2, C274H413N65O79S6 requires 6074.1. Example 9 Synthesis of NeB29-10-(4-carboxyphenoxv)-decanovl-L-alutamvlamide desB30 human insulin (Formula Removed) DesB29, DesB30 human insulin—N'^'lT—OH Step 1: 4-Hydroxbenzoic acid tert-butyl ester 4-Hydroxbenzoic acid (3 g, 21.7 mmol) was stirred in toluene (35 ml, dried over mol. sieves). The solution was heated to 80 °C under N2, and /V,A/'-dimethylformamide di-tert-butyl acetal (10.42 mL, 43.4 mmol) was added over ca. 5 min. The mixture was stirred at 80 °C for 1 h 10 min., and cooled to rt. The solution was washed with water, twice with sat. NaHC03 and sat. NaCI (15 mL each), dried over MgSO4, and concentrated to yield a yellow oil (2.77 g). The product was purified by flash chromatography (380 g silica, eluant: 4:6 AcOEt/heptane (2 L) and 1:1 AcOEt/heptane 700 mL) to yield white crystals (2.07g, 49% yield). HPLC-MS m/z: 217 (M23). 1H-NMR (CDCI3, 400 MHz) 7.90 (d, 2H), 6.85 (d, 2H), 6.10 (s, 1H), 1.59 (s, 9H). Step 2: 4-(9-Methoxycarbonvlnonvloxv) benzoic acid tert-butvl ester 4-Hydroxbenzoic acid tert-butyl ester (500 mg, 2.57 mmol) and 10-bromodecanoic acid methyl ester (683 mg, 2.57 mmol) were dissolved in acetonitrile, and K2C03 (712 mg, 5.15 mmol) was added. The mixture was refluxed under nitrogen for 16 h, and cooled to rt. The solids were filtered off, and the filtrate concentrated under vacuum. The resulting residue was dissolved in AcOEt (50 ml) and water (25 mL). The phases were separated and the organic phase was dried over MgSO4 and concentrated to yield a colorless oil (874 mg, 90% yield). HPLC-MS m/z: 402 (M23). 1H-NMR (CDCI3, 400 MHz) 7.92 (d, 2H), 6.87 (d, 2H), 3.99 (t, 2H), 3.67 (s, 3H), 2.31 (t, 2H), 1.72-1.83 (m, 2H), 1.59-1.69 (m, 2H), 1.58 (s, 9H), 1.40-1.50 (m, 2H), 1.23-1.40 (br, 8H). Step 3: 4-(9-Carboxvnonvloxy) benzoic acid tert-butvl ester 4-(9-Methoxycarbonylnonyloxy) benzoic acid tert-butyl ester (858 mg, 2.27 mmol) was dissolved in THF (5 ml), and 1N NaOH (2.27 mmol) was added. The mixture was stirred for 16 h. AcOEt (40 mL) and 1N HCI (2.38 ml) in water (25 ml) were added. The phases were separated, and the organic phase was dried over MgSO4, and concentrated under vacuum to yield a white solid (781 mg, 95% yield). HPLC-MS m/z: 387 (M23). 1H-NMR (CDCI3, 400 MHz) 7.92 (d, 2H), 6.87 (d, 2H), 3.99 (t, 2H), 2.35 (t, 2H), 1.73-1.84 (m, 2H), 1.60-1.69 (m, 2H), 1.58 (s, 9H), 1.39-1.51 (m, 2H), 1.24-1.39 (br, 8H). Step 4: 4-f9-(2,5-Dioxopvrrolidin-1-vloxvcarbonvQ nonvloxvl benzoic acid tert-butvl ester 4-(9-Carboxynonyloxy) benzoic acid tert-butyl ester (779 mg, 2.14 mmol) was dissolved in THF (15 mL), and DIEA (366 ul, 2.14 mmol) was added. The mixture was cooled to 0 °C, and placed under nitrogen, and HSTU was added. The mixture was stirred at 0 °C for 30 min and at RT for 16 h. The sample was concentrated under vacuum and AcOEt (40 ml) was added. The mixture was washed with 0.2 N HCI (2 x 25 ml), dried over MgSO4, and concentrated under vacuum to yield a slightly yellow solid. The solid was recrystallized from AcOEt to yield a white powder (276 mg, 28% yield). The mother liquor was concentrated to yield crystalline residue (430 mg, 43% yield). Data for the white powder: HPLC-MS m/z: 484 (M23). 1H-NMR (CDCI3, 300 MHz) 7.93 (d, 2H), 6.88 (d, 2H), 3.99 (t, 2H), 2.83 (s, 4H), 2.61 (t, 2H), 1.67-1.88 (m, 4H), 1.58 (s, 11H, theoret. 9H + water), 1.27-1.52 (m, 10H). Step 5: 4-[9-((S)-1-Carbamovl-3-carboxvpropylcarbamovl) nonvloxvl benzoic acid tert-butyl ester 4-[9-(2,5-Dioxopyrrolidin-1-yloxycarbonyl) nonyloxy] benzoic acid tert-butyl ester (200 mg, 0.433 mmol) was stirred in DMF (2 mL) and H-Glu-NH2 (63 mg) was added. The non-homogeneous mixture was stirred at rt for 16h. LC/MS analysis indicated the reaction had not gone to completion. H-Glu-NH2 (20 mg) and more DMF (2 mL) were added and the mixture was stirred for 2 d at rt. The sample was concentrated under vacuum and AcOEt (50 mL) was added. The solution was washed with 0.2 N HCI (2 x 25 mL) and water (25 mL), dried over MgSO4, and concentrated under vacuum to yield a white solid (180 mg, 86% yield). HPLC-MSm/z:493(M1). Step 6: 4-(9-[(S)-1-Carbamovl-3-(2,5-dioxopyrrolidin-1-vloxvcarbonvn propylcarbamoyi] nonyloxy) benzoic acid tert-butyl ester The HSTU activation was performed in manner similar to that described for 4-[9-(2,5-Dioxopyrrolidin-1-yloxycarbonyl) nonyloxy] benzoic acid tert-butyl ester. The product was purified by flash chromatography (1:1 AcOEtheptane and AcOEt) to yield 18 mg. HPLC-MSm/z:590(M1). Step 7: N£B29-10-(4-carboxvphenoxv) decanovl-L-qlutamvlamide desB30 insulin DesB30 insulin (126 mg, 0.022 mmol) was dissolved by adding 100 mM Na2C03 (1.5 mL) and acetonitrile (1.5 mL) in a 10 ml round bottom-flask. 4-{9-[(S)-1-Carbamoyl-3-(2,5-dioxopyrrolidin-1-yloxycarbonyl) propylcarbamoyi] nonyloxy} benzoic acid tert-butyl ester (14 mg, 0.022 mmol) was added in acetonitrile (750 uL) and Na2C03 (750 uL) was added so the final solution was 50:50 100 mM Na2C03/acetonitrile. The solution was stirred at RT for 1 h. The solution was transferred to a 15 ml centrifuge tube, washing with Milli-Q water (6 ml). The solution was cooled on ice, and the pH was adjusted to 5.1 by adding 1N HCI, which lead to precipitation. The tube was centrifuged at 5000 rpm for 10 min at 10 °C. The solvent was decanted from the solid. 95:5 TFA/water (2.5 ml) was added to the solid. The solution was poured into a round bottom flask, washing with more 95:5 TFA/water (2.5 ml). The solution was stirred for 30 min at RT, and concentrated under vacuum. DCM was added and removed twice, and the flask was placed under vacuum at RT. The product was purified by preparative HPLC (C18 column, acetonitrile/water/ 0.05%TFA). The relevant fractions were pooled (two batches) and diluted 1:1 with water. The solutions were cooled on ice, and the precipitation was induced by adjusting the pH to ca. 5 with 1 N NaOH. The samples were centrifuged (5000 rpm, 10 min, 5 °C). The liquid was decanted off and the pellets were lyophilized to yield a white solid (22 mg + 12 mg). MALDI-MS (alpha-cyano-4-hydroxycinnamic acid) m/z: 6128.7 (M = 6125.1). HPLC-MS m/z: 1532.8 ((M+4)/4 = 1532.2). Example 10 Synthesis of NeB29-4-[11-(4-Carboxyphenyl) undecanovlaminol butvrvl desB30 human insulin Step 1: 4-lodobenzoic acid tert-butyl ester (Formula Removed) 4-lodobenzoic acid (10 g, 40.3 mmol) was dissolved in dry toluene (100 ml, dried over mol. sieves). The solution was heated to 70 °C under a flow of nitrogen. A solution of N,N-dimethylformamide di-tert-butyl acetal (24.6 g, 121 mmol) in toluene (25 mL) was added over ca. 30 min. The reaction was mixed for 16 h. At some point the heating unit failed, so the reaction cooled from 70 °C to rt. The solution was heated to 70 °C for and mixed for 5 h. The sample was concentrated under vacuum, and AcOEt (400 ml) was added. The solution was then washed with 1:1 sat. NaHCOs/water (150 ml), and sat. NaHC03, water and sat. NaCI (75 mL each). The organic phase was dried (MgSO4) and concentrated under vacuum to yield light brown oil. HPLC-MS m/z: 327 (M23). 1H-NMR (CDCI3, 400 MHz) 5 7.77 (d, 2H), 7.69 (d, 2H), 1.58 (s, 9H). Step 2: 11-lodo undecanoic acid methyl ester 11-Bromo undecanoic acid methyl ester (20.2 g, 72.3 mmol) was dissolved in acetone (200 ml). Sodium iodide (54 g, 361 mmol) was added and reaction was refluxed under nitrogen for 16 h. After cooling to RT the salts were filtered off. The filtrate was concentrated under vacuum and water (200 ml) was added. The solution was extracted with AcOEt (2 x 100 ml) adding some sat. NaCI to aid phase separation. The organic extracts were pooled and washed with water (100 ml) plus a little sat. NaCI, and sat. NaCI (50 mL). Dry over MgSO4. The solution was a red-orange color. Three teaspoons of activated charcoal were added. After mixing, the solution was filtered through a bed of celite. The filtrate was concentrated under vacuum to yield a light yellow oil (20.96 g, 89%). HPLC-MS m/z:327(M1). 1H-NMR (CDCI3, 300 MHz) 8 3.67 (s, 3H), 3.19 (t, 2H), 2.30 (t, 2H), 1.74-1.89 (m, 2H), 1.53-1.70 (m, 2H), 1.34-1.46 (m, 2H), 1.28 (br, 10H). Step 3: 4-(10-Methoxvcarbonvldecvl) benzoic acid tert-butyl ester All glassware was dried prior to use. THF was dried over molecular sieves. LiCI was dried at 150 °C for 1 h, then stored in a closed bottle. All reaction solutions were made under nitrogen, and the solutions were transferred via syringe. 4-lodobenzoic acid tert-butyl ester (1.2 g, 3.95 mmol) was dissolved in THF (3 ml) and cooled to -30 °C. Isopropyl magnesium chloride (4.34 mmol, 2M in THF) was added over 5 minutes, and the solution was stirred for 1 hr at a temperature between -18 °C to -25°C. The solution was cooled to -22 °C, and a mixture of CuCN (0.389 g, 4.34mmol) and LiCI (0.368 g, 8.68 mmol) in THF (4.2 ml) was then added. The reaction vessel was removed from cooling and allowed to warm to RT (ca. 10 min). Trimethylphosphite (0.95 mL) was added, and after stirring for 5 min at rt, a solution of 11-iodo-undecanoic acid methyl ester (1.0 g, 3.16 mmol) in THF (3 ml) was added. The solution was mixed at rt for 16 h. Sat. NH4CI (3 ml) was added, and the solution was poured into water (60 mL). The solution was extracted with AcOEt (3 x 35 ml). The organic extracts were pooled and washed with water (30 mL) using some Sat. NaCI to aid phase separation. The solvent was removed under vacuum to yield a biphasic residue. AcOEt (ca 2 ml) was added and the flask was swirled gently. Not all of the thick white residue dissolved. The portion which dissolved was added to a column of silica (50 g) and eluted with AcOEt: heptane 1:11. The appropriate fractions were concentration under vacuum to yield an oil (1.25 g). The oil was dissolved in acetone (30 mL), and piperidine (1 mL) was added. Nal (0.8 g) was added and the mixture was stirred and refluxed for 16 h. The mixture was concentrated under vacuum and partitioned between AcOEt (50 mL) and 1 N HCI (25 ml_). The organic phase was washed with 1 N HCI (2 x 25 mL), dried over MgSO4, and concentrated under vacuum to yield a colorless oil (1.1 g). The product was purified by flash chromatography (eluant: AcOEtheptane 1:11, 150 g silica) to yield a colorless oil (0.72 g, 61%). HPLC-MS m/z: 399 (M23). 1H-NMR (CDCI3, 300 MHz) 8 7.90 (d, 2H), 7.21 (d, 2H), 3.66 (s, 3H), 2.64 (t, 2H), 2.30 (t, 2H), 1.48-1.70 (m, 13H), 1.27 (br, 12H). Step 4: 4-(10-Carboxydecvn benzoic acid tert-butyl ester The compound was prepared in analogous fashion to the procedure used in the preparation of 4-(9-Carboxynonyloxy) benzoic acid tert-butyl ester to yield a white solid (0.68 g). HPLC-MS m/z: 385 (M23). 1H-NMR (CDCI3, 300 MHz) 8 7.90 (d, 2H), 7.21 (d, 2H), 2.64 (t, 2H), 2.34 (t, 2H), 1.53-1.71 (m, 13H), 1.28 (br, 12H). Step 5: 4-f10-(2,5-Dioxopyrrolidin-1-vloxvcarbonvl)decvll benzoic acid tert-butyl ester The compound was prepared in analogous fashion to the procedure used in the preparation of 4-[9-(2,5-Dioxopyrrolidin-1-yloxycarbonyl) nonyloxy] benzoic acid tert-butyl ester. HPLC-MS m/z: 482 (M23). 1H-NMR (CDCI3, 400 MHz) 8 7.89 (d, 2H), 7.21 (d, 2H), 2.76-2.93 (m, 4H), 2.54-2.68 (m, 2H), 1.67-1.81 (M, 2H), 1.52-1.66 (m, 11H), 1.35-1.43 (M, 2H), 1.19-1.35 (br, 10H). Step 6: 4-[10-(3-Carboxv-propylcarbamoyl)decvll benzoic acid tert-butyl ester 4-[10-(2,5-Dioxopyrrolidin-1-yloxycarbonyl) decyl] benzoic acid tert-butyl ester (300 mg, 0.65 mmol) was dissolved in DMF (3 ml) and 4-amino butyric acid (67 mg, 0.65 mmol). The mixture was stirred for 16 h under nitrogen. The solvent was removed under vacuum and AcOEt (35 ml) was added. The solution was washed with 0.2 N HCI and water (15 ml each). Sat. NaHC03 was added (not intended) to the organic phase. DCM (50 ml) was added. Some of the organic phase was removed and DCM (100 ml) was added to the aqueous phase and allowed to stand overnight. The mixture was cooled on ice and the pH was adjusted to 1.9 with 4N HCI. The organic phase was isolated, dried over MgSO4 and concentrated under vacuum to yield on oil (220 mg, 76% yield). HPLC-MS m/z: 470 (M23). 1H-NMR (CDCI3, 400 MHz) 5 7.89 (d, 2H), 7.21 (d, 2H), 5.79 (br, 1H), 3.27-3.40 (m, 2H), 2.64 (t, 2H), 2.40 (t, 2H), 2.18 (t, 2H), 1.78-1.91 (m, 2H), 1.51-1.61 (m, 13H), 1.35-1.43 (M, 2H), 1.17-1.36 (br, 12H). Step 7: 4-{10-f3-(2,5-Dioxo-pvrrolidin-1-vloxvcarbonyl) propylcarbamoyll decyl) benzoic acid tert-butyl ester The compound was prepared in analogous fashion to the procedure used in the preparation of 4-[9-(2,5-Dioxopyrrolidin-1-yloxycarbonyl)nonyloxy] benzoic acid tert-butyl ester, but TSTU was used instead of HSTU. Precipitation (DCM/Heptane) yielded white crystals (180 mg, 70% yield). HPLC-MS m/z: 568 (M23). 1H-NMR (CDCI3, 400 MHz) 5 7.89 (d, 2H), 7.21 (d, 2H), 5.83 (br, 1H), 3.30-3.43 (m, 2H), 2.85 (br, 4H), 2.57-2.73 (m, 4H), 2.15 (t, 2H), 1.92-2.07 (m, 2H), 1.56-1.64 (m, 13H), 1.18-1.36 (br, 12H). Step 8: NEB29-4-H 1-(4-carboxyphenyl) undecanoylaminol butyryl desB30 human insulin The compound was prepared in analogous fashion to the procedure used in the preparation of example 9 to yield 30 mg. MALDI-MS (alpha-cyano-4-hydroxycinnamic acid) m/z: 6067 (M = 6080, reference standard (M=5706) showed M-13). HPLC-MS m/z: 1520.9 ((M+4)/4 = 1521). Example 11 Synthesis of NeB29-(3-(3-(4-[3-(7-carboxvheptanovlaminO)propoxv1butoxv)propvlcarbamov0-propionvl-v-qlutamvlamide) desB30 human Insulin (Formula Removed) Step 1: N-{3-f4-(3-tert-Butoxvcarbonvlaminopropoxv)-butoxvl-propyl)succinamic acid 123-0000-3007 1-(tert-ButoxycarbonylaminO)-4,9-dioxa-12-dodecanamine (5.0 g, 16.45 mmol) was dissolved in THF (30 mL), succinic anhydride (1.81 g, 18.1 mmol) in acetonitrile (10 mL) was added and the mixture was heated to 60 C for 4 h, and subsequently stirred at RT overnight. The mixture was evaporated to dryness and EtAc (50 mL) was added. The EtAc phase was washed with HCI (0.1 M) 3 times, dried with MgSO4and subsequently the organic phase was evaporated to dryness which gave 5.86 g (88%) of thick oil. LCMS: Rt 2.86 min; m/z (M1) 405. Calcd: 405. This product was used without further purification. Step 2: Octanedioic acid tert-butvl ester 2,5-dioxo-pyrrolidin-1-vl ester (Formula Removed) Octanedioic acid mono-tert-butyl ester (3.14 g, 13.63 mmol) was dissolved in THF (100 mL). TSTU (4.9 g, 16.3 mmol) was added and pH was adjusted to 8.5 with DIPEA (2.85 mL). The mixture was stirred under nitrogen overnight, evaporated to dryness, dissolved in EtAc (50 mL) which subsequently was extracted 2 times with HCL ( 0.1 M). The organic phase was dried with MgSO4, filtered and evaporated resulting in an slightly yellow oil (5 g, containing small amounts of solvent). LCMS: Rt6.56 min; m/z (M1) 328. Calcd: 328. Step 3: 7-(3-(4-[3-(3-carboxvpropionvlaminO)propoxvlbutoxv)propylcarbamoyl)heptanoic acid tert-butyl ester 123-0000-3012 (Formula Removed) N -{3-[4-(3-tert-Butoxycarbonylaminopropoxy)-butoxy]-propyl}succinamic acid (4.60 g, 11.37 mmol) was stirred with TFA (20 mL ) at RT for 60 min, after evaporation the residue was stripped with DCM (30 mL x2) and evaporated to dryness. The resulting oil was dissolved in acetonitrile (30 mL) and octanedioic acid tert-butyl ester 2,5-dioxo-pyrrolidin-1-yl ester ( 4.46 g, 13.6 mmol) in DMF (20 mL) was added. pH was adjusted to 8.5 with DIPEA and the mixture was stirred overnight under nitrogen. The mixture was subsequently evaporated to dryness and redissolved in EtOAc (50 mL). The EtOAc phase was extracted x3 with HCI (0.1 M), the organic layer dried over magnesium sulphate, filtered and evaporated resulting in a slightly yellow crystalline oil (6.5 g, content of solvent residues). LCMS : Rt4.31 min; m/z (M1) 517. Calcd: 517. The crude product was used for further reaction without further purification. Step 4: 7-r3-(4-(3-[3-aSV1-carbamovl-3-carboxvpropvlcarbamovnpropionvlamino1propoxv)butoxv)propvlcarbamovl]heptanoic acid tert-butyl ester 0123-0000-3078 (Formula Removed) 7-(3-{4-[3-(3-CarboxypropionylaminO)propoxy]butoxy}propylcarbamoyl)heptanoic acid tert-butyl ester (2.4 g), the crude product from step 3, was dissolved in THF (60 mL), TSTU (2,11g, 6.97 mmol) was added together with DMF (10 mL), pH was adjusted to 8.2 with DIPEA (0.8 mL). The mixture was stirred overnight under nitrogen. The mixture was evaporated and the residue dissolved in EtOAc which was extracted with HCI (0.1 M) 3 times. The organic layer was dried with magnesium sulphate, filtered and the filtrate evaporated to give 3.2 g oil. LCMS: Rt 4.57 min; m/z 614, corresponding to the activated acid. This crude product was dissolved in acetonitrile (40 mL) and L-glutamic acid amide (0.6 g, 4.1 mmol) was added together with DMF (5 mL), pH was adjusted to 8.2 with DIPEA (1.4 mL). The mixture was stirred at RT for overnight; filtration followed by evaporation afforded thick yellow oil. This was extracted between EtOAc and HCI (0.1 M) as reported above, and the resulting dried EtOAc layer gave 1.66 g crude product on evaporation. LCMS: Rt 3.62 min; m/z (M1) 645. The crude product was purified by preperative HPLC using acetonitrile/water/0.1 % TFA as eluent on C18 column (Jones, Kromasil RP18 5pm 15x225 mm). Gradient: 0.0-10.0 min 35% acetonitrile A; 10.0 - 30.0 min 35-90 % A; The product was collected in fractions from 16-18 min. The combined fractions were evaporated yielding the wanted product (1.0 g). LCMS : Rt 3.59 min; m/z (M1) 645, calcd. 645. Step 5: N£B29-(3-(3-(4-[3-(7-carboxvheptanovlaminO)propoxvlbutoxv)propvlcarbamovl)propionyl-L-glutamvlamide) desB30 human insulin 7-[3-(4-{3-[3-((S)-1-Carbamoyl-3-carboxypropylcarbamoyl)propionylamino]propoxy}butoxy)-propylcarbamoyl]heptanoic acid tert-butyl ester from step 4 (1.0 g, 1.55 mmol) was dissolved in THF (20 mL), TSTU (0.51 g, 1.7 mmol) was added and pH adjusted to >8 with DIPEA (0.27 mL). The mixture was stirred overnight under nitrogen. Evaporation followed by extraction between EtOAc and HCI (0.1 M), drying of the organic phase (MgSO4) followed by evaporation to dryness gave 21 mg oil, LCMS: Rt. 4.34 min, m/z 742. This crude product was dissolved in acetonitrile (10 ml_), pH was adjusted to 8 with Na2C03 (0.1 M) and added to a solution of desB30 human insulin (1 g) dissolved in Na2C03 solution (15 mL, pH 10.2). The mixture was stirred under nitrogen at RT for 1 h. Then pH was adjusted to 5.4 by means of HCI (2M) resulting in precipitation. The mixture was filtered, the filtrate freeze dried, and the precipitate dried in vacuum overnight. Both fractions were purified on Gilson using acetonitrile/water/0.1 % TFA as eluent on C18 column (Jones, Kromasil RP18 5um 15x225 mm). Gradient: 0.0-5.0 min 35% acetonitrile (A); 5.0 - 25.0 min 35-90 % A; The product was collected in fractions from 12-15 min. The combined fractions were evaporated, re-dissolved in water and freeze-dried yielding 27 mg of the wanted product. LCMS: Rt. 7.76 min, m/z 1570 MALDI-MS (sinnapinic acid): 6277; C28oH422N68O84S6 requires 6277. Example 12 Synthesis of NEB29-O>- carboxv-tridecanovl-amino-butanovl desB30 human insulin (Formula Removed) This compound was prepared from tetradecandioic acid and amino-butyric acid in analogy with example 1. LCMS 6032.1, C27iH407N65O79S6 requires 6032.0. Example 13 (Formula Removed) Synthesis of NeB29-(o-carboxv-undecanovl-5- amino-butanovl desB30 human insulin OH DesB29, DesB30 human insulin—N This compound was prepared from dodecandioic acid and amino-butyric acid in analogy with example 1. LCMS 6003.8, C269H403N65O79S6 requires 6004.0. Example 14 Synthesis of NEB29-w-carboxv-tetradecanovl-amino-butanovl desB30 human insulin (Formula Removed) DesB29, DesB30 human insulin —N This compound was prepared from pentadecandioic acid and Y-aminobutyric acid in analogy with example 1. LCMS 6045.6, C272H409N65O79S6 requires 6046.1. Example 15 Synthesis of NeB29-{4-M0-(4-Carboxphenoxv)-decanovlamino1-butvryl) desB30 insulin (Formula Removed) This compound was prepared from 4-(9-methoxycarbonylnonyloxy) benzoic acid tert-butyl ester in analogy with examples 9 and 10. LCMS: 6095.6, C276H409N65O79S6 requires 6094.1. Example 16 Synthesis of N£B29-{4-f(14-Carboxv-tetradecanovlaminO)-methvll-benzovl> desB30 insulin (Formula Removed) DesB29, DesB30 human insulin— N' H O This compound was prepared from 4-aminomethyl benzoic acid in analogy with example 1. LCMS: 6082.0, C275H406N66O81S6 requires 6082.1. Example 17 ieB29 Synthesis of N -16-(4-Carboxv-phenoxv)-hexadecanovl1 desB30 insulin (Formula Removed) Step 1: 16-Bromohexadecanoic acid methyl ester 16-Bromohexadecanoic acid (6 g, 17.9 mmol) was dissolved in methanol (35 mL), toluene (100 mL) and trimethylorthoformate (20 mL). Amberlyst 15 was added and the mixture was stirred under nitrogen for 16 h at 55 °C. The mixture was concentrated, and dissolved in methanol (ca. 50 ml_) and DCM (30 ml_). The resin was filtered off, and the filtrate was concentrated. The volume was increased to ca. 40 mL with methanol. Cooling produced crystals which were filtered off, washed with cold methanol and dried to yield white crystals (5.61 g, 90% yield). 1H-NMR (DMSO, 300 MHz) 3.57 (s, 3H), 3.52 (t, 2H), 2.28 (t, 2H), 1.78 (m, 2H), 1.50 (m, 2H), 1.37 (m, 2H). The remainder of the steps were performed in analogy with example 9. LCMS: 6081.2, CsysHUioN^Q1Se requires 6081.1. Example 18 Synthesis of NeB29-f4-[(15-carboxypentadecanovlaminO)benzovn- desB30 human insulin (Formula Removed) Step 1: 4-(15-tert-ButoxvcarbonvlpentadecanovlaminO)-benzoic acid Mono-tert-butyl hexadecandioate hexadecadioic acid (400 mg, 1.17 mmol) was dissolved in NMP (6 ml). 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (223 mg, 1.17 mmol) and 1-hydrox7-azabenzotriazole (156 mg, 1.17 mmol) was added and the mixture was heated to 50 degrees celcius for 90 min. 4-aminobenzoic acid (320 mg, 2.34 mmol) and DIEA (0.6 mL, 3.51 mmol) was added and the mixture was stirred under nitrogenflow overnight. The mixture is transferred is separated between saturated aqueous NaHCO3(50 mL) and diethyl ether (100 mL). NaHSO4 (50 mL, 10 % in water) was added and the organic phase isolated, dried (MgSO4) and solvent removed in vacuo. The crude product was recystallized from ethanol to give 4-(15-tert-butoxycarbonylpentadecanoylaminO)-benzoicacid. 1H-NMR (DMSO-de) 8: 10.18 (s, 1H), 7.87 (d, 2H), 7.69 (d, 2H), 2.33 (t, 2H), 2.15 (t, 2H), 1.58 (t, 2H), 1.46 (t, 2H), 1.38 (s, 9H)m 1.31-1.20 (m, 20H). Step 2: 4-(15-tert-Butoxvcarbonvl-pentadecanovlaminO)-benzoic acid 2,5-dioxo-pyrrolidin-1-vl ester 4-(15-tert-ButoxycarbonylpentadecanoylaminO)-benzoic acid (29 mg, 0.063 mmol) was converted to the succinimidyl ester using TSTU similar to described above. HPLC-MS, m/z = 559 (M23, M+Na). Step 3: NEB29-(4-[(15-carboxypentadecanovlaminO)benzovll- desB30 human insulin DesB30 human insulin (355 mg, 0.062 mmol) was dissolved in DMSO (3.5 mL), triethylamin (0.622 mmol, 0.087 mL) was added. 4-(15-tert-Butoxycarbonyl-pentadecanoylaminO)-benzoic acid 2,5-dioxo-pyrrolidin-1-yl ester (28.3 mg, 0.051 mmol) was dissolved in DMSO (0.5 mL and added). The mixture was carefully stirred for 30 minutes at room temperature. The mixture was frozen by cooling with an icebath and water (7 mL) was added and the mixture left standing at room temperature until the frozen mixture had dissolved. pH was adjusted to pH = 5.3 using 1 N HCI and precipitate isolated by centrifuge, washed once with water followed by centrifuge. Trifluoroacetic acid (15 mL) was added and the mixture was stirred for 30 minutes, poured into diethyl ether (50 mL) while maintaining cooling on an icebath. The crude product was isolated by centrifuge and dissolved in 10 mM TRIS + 15 mM (NH4)2SO4in 20 % EtOH, pH 7.3 and subjected to purification on an AKTA purifier employing a reversed phase HPLC, Jupiter 5269, C4 250/20 mm, 15 nM, 300 A. The buffer consisted of A-buffer 10 mM TRIS + 15 mM (NH4)2SO4in 20 % EtOH, pH 7.3 and a ß-buffer 80 % EtOH. The product was eluted with a gradient 15-60 % B with 8 ml/min. Fractions containing product were collected an pH adjusted to pH = 5.2. NEB29-{4-[(15-carboxypentadecanoylaminO)benzoyl]- desB30 human insulin was isolated by centrifuges and lyophilized. LCMS: 6092.0, C276H409N65O79S6 requires 6094.1. Example 19 Synthesis of N£B29-{4-f(15-Carboxv-pentadecanoylami nO)-methvll-benzoylVdesB30 insulin (Formula Removed) Step 1: 4-[(15-tert-ButoxvcarbonvlpentadecanovlaminO)methvHbenzoic acid Hexadecanedioic acid tert-butyl ester 2,5-dioxo-pyrrolidin-1-yl ester (370 mg, 0.842 mmol) was dissolved in NMP (8 mL), 4 aminomethyl benzoic acid (127,2 mg, 0.842 mmol) was added , the mixture was stirred at room temperature overnight, followed by heating at for 2 hours at 50 degrees celcius. After cooling to room temperature the mixture was poured into water. Product isolated by filtration dried and used for the next step without any further purification. 1H-NMR (DMSO-d6) (selected signals) 8: 7.88 (d, 2H), 7.34 (d, 2H), 4.31 (d, 2H), 2.15 (m,4H), 1.47 (m,4H), 1.38 (s,9H), 1.23 (br s, 2H), 1.38 (s, 9H). HPLC-MS: m/z = 498 (M+Na) Step 2. NEB29-{4-[(15-Carboxv-pentadecanoylami nO)-methvn-benzovl)-desB30 insulin The 4-[(15-tert-ButoxycarbonylpentadecanoylaminO)methyl]benzoic acid was converted to converted to the succinimidyl ester and reacted with DesB30 human insulin similar to described above. The product was purified on an AKTA purifier employing a reversed phase HPLC, Jupiter 5269, C4 250/20 mm, 15 u,M, 300 A. The buffer consisted of A-buffer 10 mM TRIS + 15 mM (NH4)2SO4in 20 % EtOH, pH 7.3 and a ß-buffer 80 % EtOH. The product was eluted with a gradient 15-60 % B with 8 ml/min. Followed by purification with Rß-HPLC on a Waters Prep LC2000, on C18, 5 cmx20 cm, flow 20 ml/min using acetonitrile/water 36-42 % gradient containing 0.1 % TFA. Fraktions containing product was collected and lyophilized. To the lyophilized material was added water (7.2 ml_) and pH adjusted to 8.98 with 1 N+0.1 N NaOH. The pH was adjusted back to 5.2-5.5 with 0.1 N HCI. The product precipitated, isolated by centrifuge and lyophilized to give the title compound. LCMS: 6108.0, C277H411N65O79S6 6108.1 Example 20: Hydrophobicitv. albumin affinity, self-association and mixability of long.acting and short-acting insulins Analysis of selv-associating properties of the insulin derivatives of the invention The ability of the insulin derivatives of the invention to self-associate into large, but soluble complexes is analysed using SEC (size exclusion chromatoghaphy): Column: Superose™ 6 PC 3.2/30, CV = 2.4 ml (Amerham Biosciences) Temperature: 37 °C SEC buffer: 140 mM NaCI, 10 mM TrisHCI, 0.01% NaN3, pH 7.5 Injection volume: 20 pi Flow: 0.05 ml/min Runtime: 60 min and equillibration of additional 100 min For this analysis the insulin derivatives of the invention are in a solution consisting of 0.6 mM derivative, 2.1 Zn2+/hexamer, 16 mM phenol, 7 mM phosphate pH 7.8. The retention time of the derivative is then compared to the retention times of the following standard molecules: Blue dextran (> 5 MDa, KAV 0.O), Thyroglobulin (669 kDa, KAV 0.28), Ferritin (440 kDa, KAV 0.39), Ovalbumin (44.5 kDa, KAv 0.56), Ribonuclease (13.7 kDa, KAv 0.69) and a second reference of Albumin (66 kDa, KAv0.53), Co(lll)1-nsulin-hexamer (35 kDa, KAV0.61), and monomeric insulin X2 (6 kDa, KAV 0.73). The following equation is used to determine the Kav for the derivative: Kav=(t-tO)/(Vt/(f+td-tO)) Where t is the retention time for a given peak, to is the retention time for Blue dextran, Vt is the total column volume (here 2.4 ml), f is the flow (here 0.04 ml/min), and td is the retention time for Blue dextran without the column in the system. The Kav value indicates the degree of selv-association of a derivative, i.e. a large KaV similar to the Kav for the Co(lll)1-nsulin hexamer and X2 insulin monomer shows low or no propensity of the derivative to form large, selv-associated complexes, while very small Kav close to zero or even negative shows great propensity of the derivative for selv-association into large, soluble complexes. Hvdrophobicitv data on insulin derivatives according to the invention. The hydrophobicity (hydrophobic index) of the insulin derivatives of the invention relative to human insulin, k'rei, was measured on a LiChrosorb RP18 (5um, 250x4 mm) HPLC column by isocratic elution at 40 °C using mixtures of A) 0.1 M sodium phosphate buffer, pH 7.3, containing 10% acetonitrile, and B) 50% acetonitrile in water as eluents. The elution was monitored by following the UV absorption of the eluate at 214 nm. Void time, to, was found by injecting 0.1 mM sodium nitrate. Retention time for human insulin, thUman, was adjusted to at least 2to by varying the ratio between the A and B solutions. k'rei= (tderivative-tO)/(thuman-tO)- k'rei found for a number of insulin derivatives according to the invention are given in Table 1. Human Serum Albumin Affinity Assay Relative binding constant of 125l-TyrA14-analogue to human serum albumin immobilised on Minileak particles and measured at 23 °C (detemir =1 in saline buffer). Mixability of lonacting and short-acting insulins as analyzed by size-exclusion chromatography of insulin mixtures SEC: Mixability of Insulin Aspart (3 Zn/6 insulin, glycerol 1.6%, 16 mM phenol and 16 mM tr1-6-cresol, sodium chloride 10 mM, phosphate 7 mM, pH 7.4) and prolonged acting insulin (2.1 or 6 Zn/6 insulin) 30:70, as measured by collecting fractions from SEC (as described above) and quantifying by HPLC the presence of prolonged-acting and fast-acting insulins in the high molecular weight fraction (fraction 2, MW > HSA) and in the low molecular weight fraction (fraction 3, MW = HSA), respectively. Four fractions are collected at size of 16 min after delay, of which fraction 2 [16-32 min] (peak 1) contain associated form larger than albumin (32 min correspond to KAvn 0.46) and fraction 3 (peak 2) contain dihexameric, hexameric, dimeric and monomeric forms of insulin. HPLC: Reverse phase chromatography on a Zorbax Eclipse XDß-C18 2.1*15 mm (1.8 urn) gradient eluted with buffer A: 0.2 M sodium sulphate, 0.04 M sodium phosphate, 10 % acetonitrile, pH 7.2 and buffer B: 70 % acetonitrile at 30 °C, 19-34 % B in 4.5 min. linear, sudden initial condition at 5 min., run time of 7 min., flow of 0.5 ml/min., injection volumen of 14 uL and UV detection at 276 nm using Insulin Aspart reference of 609 uM for both analogues. (Table Removed) Table legend: Hydrophobicity relative to human insulin: k'rel <1: +++, 1-10: ++, >10: + (HI =1) Insulin receptor affinity relative to human insulin: <5%: +, 5-50%: ++, >50%: +++ Human serum albumin affinity relative to insulin detemir: <0.5: +, 0.5-2: ++, > 2: +++ Self-association: Kav < 0.1: +++, Kav <0.55: ++ and Kav >0.55: + Kav = 0.55 for human serum albumin, Kav = 0.63 for human insulin Co(lll)hexamer, Kav = 0.72 for the monomeric insulin analogue X2. n.a. = not analyzed. Example 21: Euglvcaemic glucose clamp after s.c. administration of insulin preparations to pigs Female pigs, 60-90 kg, fasted for 18 h. During the experiments the pigs are free to move in their pens. An insulin dose is administered s.c, depending of dose size often divided in two depots. Each pig is kept euglycaemic at its individual fasting glucose levels for up to 24 h by a variable rate intravenous infusion of a 20 % glucose solution. The infusion is given through a catheter inserted in the jugular vein. Depending on changes in plasma glucose concentrations observed during frequent plasma glucose monitoring, the necessary adjustments of the glucose infusion are made empirically. Blood samples are collected in EDTA glass tubes every 15-30 min, plasma separated for glucose and insulin measurements. Glucose is determined within 1.5 min of blood sampling with an YSI (Yellow Springs Instruments) glucose analyser (glucose oxidase method). Mean glucose infusion rate (GIR) profiles and mean plasma insulin profiles are made for each insulin preparation. (Figures 2-4 show mean ± SEM). PHARMACOLOGICAL METHODS Assay (I) Insulin receptor binding of the insulin derivatives of the invention The affinity of the insulin analogues of the invention for the human insulin receptor was determined by a SPA assay (Scintillation Proximity Assay) microtiterplate antibody capture assay. SPA-ßVT antibodbinding beads, anti-mouse reagent (Amersham Biosciences, Cat No. PRNQ0017) were mixed with 25 ml of binding buffer (100 mM HEPES pH 7.8; 100 mM sodium chloride, 10 mM MgSO4, 0.025% Tween-2O). Reagent mix for a single Packard Optiplate (Packard No. 600519O) is composed of 2.4 ul of a 1:5000 diluted purified recombinant human insulin receptor - exon 11, an amount of a stock solution of A14 Tyr[125l]-human insulin corresponding to 5000 cpm per 100 ul of reagent mix, 12 ul of a 1:1000 dilution of F12 antibody, 3 ml of SPA-beads and binding buffer to a total of 12 ml. A total of 100 ul was then added and a dilution series is made from appropriate samples. To the dilution series was then added 100 ul of reagent mix and the samples were incubated for 16 hours while gently shaken. The phases were the then separated by centrifugation for 1 min and the plates counted in a Topcounter. The binding data were fitted using the nonlinear regression algorithm in the GraphPad Prism 2.01 (GraphPad Software, San Diego, CA). Assay (II) Potency of the insulin derivatives of the invention relative to human insulin Sprague Dawley male rats weighing 238-383 g on the experimental day are used for the clamp experiment. The rats has free access to feed under controlled ambient conditions and fast overnight (from 3 pm) prior to the clamp experiment. Experimental Protocol The rats are acclimatized in the animal facilities for at least 1 week prior to the surgical procedure. Approximately 1 week prior to the clamp experiment Tygon catheters are inserted under halothane anaesthesia into the jugular vein (for infusion) and the carotid artery (for blood sampling) and exteriorised and fixed on the back of the neck. The rats are given Streptocilin vet. (Boehringer Ingelheim; 0.15 ml/rat, i.m.) post-surgically and placed in an animal care unit (25 °C) during the recovery period. In order to obtain analgesia, Anorphin (0.06 mg/rat, s.c.) is administered during anaesthesia and Rimadyl (1.5 mg/kg, s.c.) is administered after full recovery from the anaesthesia (2-3 h) and again once daily for 2 days. The clamp technique employed is adapted from (1). At 7 am on the experimental day overnight fasted (from 3 pm the previous day) rats are weighed and connected to the sampling syringes and infusion system (Harvard 22 Basic pumps, Harvard, and Perfectum Hypodermic glass syringe, Aldrich) and then placed into individual clamp cages where they rest for ca. 45 min before start of experiment. The rats are able to move freely on their usual bedding during the entire experiment and had free access to drinking water. After a 30 min basal period during which plasma glucose levels were measured at 10 min intervals, the insulin derivative to be tested and human insulin (one dose level per rat, n = 6-7 per dose level) were infused (i.v.) at a constant rate for 300 min. Plasma glucose levels are measured at 10 min intervals throughout and infusion of 20% aqueous glucose is adjusted accordingly in order to maintain euglyceamia. Samples of re-suspended erythrocytes were pooled from each rat and returned in about 1/4 ml volumes via the carotid catheter. On each experimental day, samples of the solutions of the individual insulin derivatives to be tested and the human insulin solution are taken before and at the end of the clamp experiments and the concentrations of the peptides were confirmed by HPLC. Plasma concentrations of rat insulin and C-peptide as well as of the insulin derivative to be tested and human insulin are measured at relevant time points before and at the end of the studies. Rats are killed at the end of experiment using a pentobarbital overdose. Test compounds and doses: Insulins to be tested are diluted from a stock solution containing 97 uM of the insulin derivative in 5mM phosphate pH 7.7. The final concentration in the solution ready for use is 0.45 uM of the insulin derivative, 5 mM of phosphate, 100 mM of sodium chloride, 0.007% of polysorbate 20. The pH was 7.7 and the i.v. infusion rate was 15 and 20 pmolmin"1kg~1. A stock solution of human insulin that is used as reference compound was formulated in a similar medium and infused i.v. at 6, 15 or 30 pmolmin"1 kg"1. Both stock solutions are stored at -20 °C and thawed overnight at 4 °C before use. The solutions are gently turned upside down several times 15 min before they are transferred to the infusion syringes. Assay (III) Determination in pigs of T50% of the insulin derivatives of the invention T50% is the time when 50% of an injected amount of the A14 Tyr[125l] labelled derivative of an insulin to be tested has disappeared from the injection site as measured with an external v-counter. The principles of laboratory animal care are followed, Specific pathogen-free LYYD, non-diabetic female pigs, cross-breed of Danish Landrace, Yorkshire and Duroc, are used (Holmenlund, Haarloev, Denmark) for pharmacokinetic and pharmacodynamic studies. The pigs are conscious, 4-5 months of age and weighing 70-95 kg. The animals fast overnight for 18 h before the experiment. Formulated preparations of insulin derivatives labelled in Tyr*14 with 125l are injected sc. in pigs as previously described (Ribel, U., Jorgensen, K, Brange, J, and Henriksen, U. The pig as a model for subcutaneous insulin absorption in man. Serrano-Rios, M and Lefebvre, P. J. 891-896. 1985. Amsterdam; New York; Oxford, Elsevier Science Publishers. 1985 (Conference Proceeding)). At the beginning of the experiments a dose of 60 nmol of the insulin derivative according to the invention (test compound) and a dose of 60 nmol of insulin detemir (both 125l labelled in Tyr A14) are injected at two separate sites in the neck of each pig. The disappearance of the radioactive label from the site of sc. injection is monitored using a modification of the traditional external gamma-counting method (Ribel, U. Subcutaneous absorption of insulin analogues. Berger, M. and Gries, F. A. 70-77 (1993). Stuttgart; New York, Georg Thime Verlag (Conference Proceeding)). With this modified method it is possible to measure continuously the disappearance of radioactivity from a subcutaneous depot for several days using cordless portable device (Scancys Laboratorieteknik, Vaerl0se, DK-3500, Denmark). The measurements are performed at 1-min intervals, and the counted values are corrected for background activity. CLAIMS 1. An insulin derivative having a formula (Formula Removed) wherein Ins is a parent insulin moiety and Q1—Q2—[CH2]n—X1—[CH2]n —Q3—[CH2]n —X2—[CH2]n —Q4—[CH2]n —X3—[CH2]n —Q5—[CH2]n —Z is a substituent and where the Ins is attached to the substituent via an amide bond between the a-amino group of the N-terminal amino acid residue of the B chain of Ins or an e-amino group of a Lys residue present in the A or B chain of Ins and a CO group in Q1 or Q2 of the substituent; each n is independently 0,1,2, 3, 4, 5 or 6; Q1 is: • an amino acid amide of an amino acid with a carboxylic acid in the side chain, or an amino acid with an uncharged side chain, which residue forms, with its carboxylic acid group, an amide group together with the a-amino group of the N-terminal amino acid residue of the B chain of Ins or together with the -amino group of a Lys residue present in the A or B chain of Ins, or • a chain composed of two, three or four a-amino acid amide or amino acid residues as specified above linked together via amide bonds, which chain - via an amide bond - is linked to the α-amino group of the N-terminal amino acid residue of the B chain of Ins or to the -amino group of a Lys residue present in the A or B chain of Ins, or • a bond Q2 is: -COCH(CONH2)- -COCH2N(CH2CONH2)- -COCH2N(CH2CONH2)COCH2N(CH2CONH2) -COCH2CH2N(CH2CH2CONH2)- -COCH2CH2N(CH2CH2CONH2)-COCH2CH2N(CH2CH2CONH2)- -COCH2N(CH2CH2CONH2)- -COCH2CH2N(CH2CONH2)- • -COCH2OCH2CONH- -CO-((CR5R6)1-6 -NH-CO)1-4-; -CO-((CR5R6)1-6 -CO-NH) 1-4_ where R5 independently can be H, —CH3,—(CH2)1-6CH3 or —CONH2 and R6 independently can be H, —CH3,—(CH2)1-6CH3; or • a bond provided that at least one of Q1 or Q2 is not a bond, and - that Q2 is not -CO-(CH2)2 -CO-NH- when n is 0 or 1, X1 is a bond and Q3 is (CH2CH2O)2- (CH2CH2O)3- or (CH2CH2OCH2CH2CH2CH2O)- and that if an amine in Q1 or Q2 forms a bond with the rest of the substituent, the amine must be bound to the rest of the substituent via a carbonyl group; Q3, Q4, and Q5 independently of each other can be • -(CH2)m- where m is an integer in the range of 6 to 32; • a divalent hydrocarbon chain comprising 1, 2 or 3 -CH=CH- groups and a number of -CH2- groups sufficient to give a total number of carbon atoms in the chain in the range of 4 to 32; -CO-((CR5R6)1-6 -NH-CO)-; -(CO-(CR5R6)1-6-CO-NH) 1-4, where R5 independently can be H, — CH3,—(CH2)1-6CH3 or —CONH2 and R6 independently can be H, — CH3,—(CH2)1-6CH3; • -CO-(CH2) 0-3-Ar-(CH2) 0-3- where Ar can be arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of -CH3, -(CH1-6-CH3, -CONR1R2 or-SO2NR1R2, where R1 and R2, independently of each other can be H, -CH3 or - (CH)1-6-CH3; • (CH2CH2O)y-; (CH2CH2CH2O)y-; (CH2CH2CH2CH2O)y-; (CH2CH2OCH2CH2CH2CH2O) y- or (CH2CH2CH2OCH2CH2CH2CH2O)y-; -(CH2OCH2)y- where y is 1-20; • arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of -CH3, -(CH)1-6-CH3, -CONR1R2 or-SO2NR1R2, where R1 and R2, independently of each other can be H, -CH3 or -(CH)1.6-CH3; • a chain of the formula -(CH2)s-Y1-(Ar)v1 -Y2- (CH2)W-Y3- (Ar)v2-Y4- (CH2)t-Y5- (Ar)v3-Y6- (CH2)Z-wherein Ar is defined as above, Y1 - Y6 independently of each other can be O, S, S=0, SO2 or a bond; where s, w, t and z independently of each other are zero or an integer from 1 to 10 so that the sum of s, w, t and z is in the range from 4 to 30, and v1, v2, and v3 independently of each other can be zero or 1 with the proviso that Y1 - Y6 do not link to each other and that the structure -O-(CH2)1-O- does not occur; or • a bond; with the proviso that at least one of Q3 - Q5 is not a bond; X1 , X2 and X3 are independently of each other O; -c=o • a bond; NCOR1, where R1 can be H, -CH3 or -(CH)1-6-CH3; or (Formula Removed) where R is hydrogen, C1-3-alkyl, C2-3-alkenyl or C2-3-alkynyl; with the proviso that X1 , X2 and X3 cannot bind to Z and - when X! , X2 and X3are O, then X1 , X2 and X3 do not bind directly to O in Q3, Q4, and Q5 and Zis: -COOH; -CO-Asp; -CO-Glu; -CO-Gly; -CO-Sar; -CH(COOH)2; -N(CH2COOH)2; -SO3H -OSO3H -OP03H2 -PO3H2 or -tetrazol-5-yl or -O-Wi, where Wi is arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of tetrazo-5-lyl, -COOH, -SO3H, -(CH2)1-6-SO3H, -(CH2)1. 6—O—PO3H2,-CONR3R4 or-SO2NR3R4, where R3and R4, independently of each other can be H,-(CH2)1-6-SO3H, or -(CH2)1-6—O—PO3H2; provided that when Z is -O-W1 then Q1 must be present and any Zn2+ complex thereof. 2. Insulin derivative according to any of claim 1, wherein Q2 is selected from the group consisting of -COCH(CONH2)- -COCH2N(CH2CONH2)- -COCH2N(CH2CONH2)COCH2N(CH2CONH2) • -COCH2CH2N(CH2CH2CONH2)- • -COCH2CH2N(CH2CH2CONH2)-COCH2CH2N(CH2CH2CONH2)--COCH2N(CH2CH2CONH2)- • -COCH2CH2N(CH2CONH2)--COCH2OCH2CONH-; -CO-((CR5R6)1-6-NH-CO)i-*-;or -CO-((CR5R6)1-6 -CO-NH) 1-4-, where R5 independently can be H, — CH3)—(CH2)1-6CH3 or —CONH2 and R6 independently can be H, —CH3l—(CH2)1-6CH3 3. Insulin derivative according to any of paragraphs 1-9, wherein Q3 is -(CH2)m-where m is an integer in the range of 6 to 32 or from 8 to 20 or m is 12, 13, 14, 15 or 16. 4. Insulin derivative according to claims 1-3 wherein Q1 Q4, Q5, X1 X2 and X3 is bonds and n is zero. 5. Insulin derivative according to claim 1, wherein one of Q3, Q4, or Q5 is (CH2CH2O)y-; (CH2CH2CH2O) y-; (CH2CH2CH2CH2O)y-; (CH2CH2OCH2CH2CH2CH2O)y- or (CH2CH2CH2OCH2CH2CH2CH2O)y-; -(CH2OCH2)y- where y is 1-20. 6. Insulin derivative according to claim 1, wherein Q3 is -CO-((CR5R6)1-6 -NH-CO)-; -(CO-(CR5R6)1-6 -CO-NH) 1-4 where R5 independently can be H, — CH3l—(CH2)1 6CH3 or —CONH2 and R6 independently can be H, —CH3l—(CH2)i-6CH3; • -CO—(CH2) 0-3-Ar-(CH2) 0-3- where Ar can be arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of -CH3, -(CH)1-6-CH3, -CONR1R2 or -SO2NR1R2, where R1 and R2, independently of each other can be H, -CH3 or -(CH)1-6-CH3; or • a bond Q4 is • -(CH2)m- where m is an integer from 4 to 22; • a divalent hydrocarbon chain comprising 1, 2 or 3 -CH=CH- groups and a number of -CH2- groups sufficient to give a total number of carbon atoms in the chain in the range of 4 to 22; • arylene or heteroarylene, which may be substituted with one or two groups selected from the group consisting of -CH3, -(CH)1-6-CH3, -CONR1R2 or-SO2NR1R2, where R1 and R2, independently of each other can be H, -CH3 or -(CH1-6-CH3; or • a chain of the formula -(CH2)s-Y1(Ar)v1 -Y2- (CH2)W-Y3- (Ar)v2-Y4- (CH2)t-Y5- (Ar)v3-Y6- (CH2)Z-wherein Ar is defined as above, Y1 - Y6 independently of each other can be O, S, S=O, SO2 or a bond; where s, w, t and z independently of each other are zero or an integer from 1 to 10 so that the sum of s, w, t and z is in the range from 4 to 30, and v1, v2, and v3 independently of each other can be zero or 1 with the proviso that Y1 - Y6do not link to each other and that the structure -O-(CH2)1-O- does not occur; X1 is •O; • -c=o • NCOR1, where R1 can be H, -CH3 or -(CH)1-6-CH3; or where R is hydrogen, C1-3-alkyl, C2.3-alkenyl or C2-3-alkynyl; with the proviso that when X1 is O, then X1 does not bind directly to O in Q4; X2, X3 and Q5 are bonds; All values of n are zero; and Z is: -COOH; -CO-Asp; -CO-Glu; -CO-Gly; -CO-Sar; -CH(COOH)2; -N(CH2COOH)2; -SO3H -OSO3H -OPO3H2 -PO3H2 or -tetrazol-5-yl or -O-W1 where W1 is arylene or heteroaryiene, which may be substituted with one or two groups selected from the group consisting of tetrazo-5-lyl, -COOH, -SO3H, -(CH2)1.6-SO3H, -(CH2)1. 6—O—PO3H2,-CONR3R4 or-SO2NR3R4, where R3 and R4, independently of each other can be H,-(CH2)1-6-SO3H, or -(CH2)1-6—O—PO3H2; and any Zn2+ complex thereof. 7. Insulin derivative according to any of paragraphs 1-6 wherein Z is -COOH. 8. An insulin derivative according to any of the claims 1-7, wherein the parent insulin is an insulin analogue. 9. Insulin derivative according to claim 8, wherein the parent insulin is selected from the group consisting of: desB30 human insulin, GlyA21 human insulin, GlyA21desB30 human insulin, GlyA21ArgB31ArgB32 human insulin, LysB3GluB29 human insulin, LysB28ProB29 human insulin and ThrB29LysB30 human insulin. 10. A pharmaceutical composition for the treatment of diabetes in a patient in need of such treatment, comprising a therapeutically effective amount of an insulin derivative according to any of the preceding claims together with a pharmaceutically acceptable carrier. 11. A method for producing a pharmaceutical composition according to claim 10, wherein up to about 10 zinc atoms per 6 molecules of insulin derivative are added to the pharmaceutical composition. 12. A method of treating diabetes in a patient in need of such a treatment, comprising administering to the patient a therapeutically effective amount of an insulin derivative according to any of claims 1-10. 13. A method according to claim 12 for pulmonary treatment of diabetes. 14. A mixture of an insulin derivative according to any of claims 1-9 and a rapid acting insulin analogue selected from the group consisting of AspB28 human insulin; LysB28ProB29 human insulin and LysB3GluB29 human insulin. 15. An insulin derivative, wherein the insulin derivative is selected from the group consisting of: NeB29--carboxy-pentadecanoyl--L-glutamylamide desB30 human insulin, NeB29--carboxy-pentadecanoyl--amino-butanoyl desB30 human insulin, NeB29--carboxy-tetradecanoyl-L-glutamylamide desB30 human insulin, NeB29--carboxy-tridecanoyl--L-glutamylamide desB30 human insulin, NEB29--carboxy-pentadecanoyl-ß-alanyl desB30 human insulin, NeB29--carboxy-pentadecanoyl--L-aspartylamide desB30 human insulin, NEB29--carboxy-pentadecanoyl--aminohexanoyl desB30 human insulin, NeB29--carboxy-pentadecanoyl-δ-aminopentanoyl desB30 human insulin, NEB29-10-(4-carboxyphenoxy)-decanoyl--L-glutamylamide desB30 human insulin, NeB29-4-[11-(4-Carboxyphenyl) undecanoyiamino] butyryl desB30 human insulin, NeB29-(3-(3-{4-[3-(7-carboxyheptanoylamino)propoxy]butoxy}propylcarbamoyl)-propionyl-y-glutamylamide) desB30 human Insulin, NeB29-- carboxy-tridecanoyl--amino-butanoyl desB30 human insulin, NeB29-)-carboxy-undecanoyl-- amino-butanoyl desB30 human insulin, NEB29--carboxy-tetradecanoyl--amino-butanoyl desB30 human insulin, NeB29-{4-[10-(4-Carboxy-phenoxy)-decanoylamino]-butyryl} desB30 insulin, NEB29-{4-[(14-Carboxy-tetradecanoylamino)-methyl]-benzoyl}desB30 insulin, NB29-[16-(4-Carboxy-phenoxy)-hexadecanoyl] desB30 insulin, NB29-{4-[(15-carboxypentadecanoylamino)benzoyl]- desB30 human insulin and NeB29-{4-[(15-Carboxy-pentadecanoylamino)-methyl]-benzoyl}-desB30 insulin